The fossil record is notorious for being incomplete and distorted, frequently conditioning the type of knowledge that can be extracted from it. In many cases, this often leads to issues when performing complex statistical analyses, such as classification tasks, predictive modelling, and variance analyses, such as those used in Geometric Morphometrics. Here different Generative Adversarial Network architectures are experimented with, testing the effects of sample size and domain dimensionality on model performance. For model evaluation, robust statistical methods were used. Each of the algorithms were observed to produce realistic data. Generative Adversarial Networks using different loss functions produced multidimensional synthetic data significantly equivalent to the original training data. Conditional Generative Adversarial Networks were not as successful. The methods proposed are likely to reduce the impact of sample size and bias on a number of statistical learning applications. While Generative Adversarial Networks are not the solution to all sample-size related issues, combined with other pre-processing steps these limitations may be overcome. This presents a valuable means of augmenting geometric morphometric datasets for greater predictive visualization.

The rapid increase in unstructured data across various fields has made multi-document comprehension and summarization a critical task. Traditional approaches often fail to capture relevant context, maintain logical consistency, and extract essential information from lengthy documents. This paper explores the use of Long-context Large Language Models (LLMs) for multi-document summarization, demonstrating their exceptional capacity to grasp extensive connections, provide cohesive summaries, and adapt to various industry domains and integration with enterprise applications/systems. The paper discusses the workflow of multi-document summarization for effectively deploying long-context LLMs, supported by case studies in legal applications, enterprise functions such as HR, finance, and sourcing, as well as in the medical and news domains. These case studies show notable enhancements in both efficiency and accuracy. Technical obstacles, such as dataset diversity, model scalability, and ethical considerations like bias mitigation and factual accuracy, are carefully analyzed. Prospective research avenues are suggested to augment the functionalities and applications of long-context LLMs, establishing them as pivotal tools for transforming information processing across diverse sectors and enterprise applications.

Recently, there has been considerable research on deepfake detection. However, most existing methods face challenges in adapting to the advancements in new generative models within unknown domains. In addition, the emergence of new generative models capable of producing and editing high-quality images, such as Diffusion, Consistency, and LCM, poses a challenge for traditional deepfake training models. These advancements highlight the need for adapting and evolving existing deepfake detection techniques to effectively counter the threats posed by sophisticated image manipulation technologies. In this paper, our objective is to detect deepfake videos in unknown domains using unlabeled data. Specifically, our proposed approach employs Meta Pseudo Labels (MPL) with Supervised Contrastive Learning, so-called SupCon-MPL, allowing the model to be trained on unlabeled images. MPL involves the simultaneous training of both a Teacher model and a Student model, where the Teacher model generates Pseudo Labels utilized to train the Student model. This method aims to enhance the adaptability and robustness of deepfake detection systems against emerging unknown domains. Supervised Contrastive Learning utilizes labels to compare samples within similar classes more intensively, while encouraging greater distinction from samples in dissimilar classes. This facilitates the learning of features in a diverse set of deepfake images by the model, consequently contributing to the performance of deepfake detection in unknown domains. When utilizing the ResNet50 model as the backbone, SupCon-MPL exhibited an improvement of 1.58% in accuracy compared to traditional MPL in known domain detection. Moreover, in the same generation of unknown domain detection, there was a 1.32% accuracy enhancement, while in the detection of post-generation unknown domains, there was an 8.74% increase in accuracy.

Binarization of degraded documents is an important preprocessing task for various document analysis such as OCR and historical document analysis. Existing studies have applied various convolutional neural network (CNN) models and generative models for document binarization, but they do not show generalized performance for noise that the model has not seen and it suffers from extracting elaborate text strokes. In this paper, to overcome these challenges, we utilize latent diffusion model (LDM), which is known for high-quality image generation model, for the first time in document binarization. By utilizing the iterative diffusion-denoising process in latent space, it shows high-quality cleaned binarized image generation and high generalized performance through using both data distribution and time step while training. Additionally, we apply gated U-Net to the backbone network to preserve text strokes using trainable gating value. Gated convolution can extract elaborate text stroke by allowing the model to focus on text region by combining gating value and feature. Furthermore, we maximize the effectiveness of the proposed model by training it with a combination of LDM loss and pixel-level loss, which is suitable for the model structure. Experiments on H-DIBCO and DIBCO benchmark datasets show that the proposed model outperforms existing methods.

Current stock market forecasting methods encompass fundamental, technical, emotional, and bargaining factors. Predominantly, price prediction hinges on order volume and price, although correlating these two within existing models proves challenging. This study employs Cycle Generative Adversarial Network (Cycle GAN) to unravel the intricate price-volume relationship, combining it with Bollinger Bands for trading signal analysis, overcoming hurdles in short-term forecasting prevalent in numerical analysis and AI. Focusing on TSMC (2330.TW) stock price, the research leverages Cycle GAN in deep learning to master the price-volume nexus, juxtaposed with LSTM and RNN. Historical TSMC closing prices and transaction counts are model inputs, scrutinizing their interconnectedness for predictions. This innovative approach aligns stock price, volume, market value, taxes, and prior changes via system engineering. By intertwining Bollinger Bands with stock price forecasts, trading signals are distilled, factoring in extended index %b for a comprehensive market picture.

Aiming at the difficult problem of broadband oscillation localization in power systems, the intel-ligent localization method of oscillation source based on the digital twin is proposed and the os-cillation source localization system based on the digital twin is constructed. Firstly, a digital twin-based oscillation source localization method and system architecture are proposed. Fur-thermore, an intelligent positioning method of oscillation source based on data-driven and mechanism fusion is proposed. It includes three steps: oscillation signal preprocessing, oscillation modal analysis and oscillation source localization. For the oscillation signal preprocessing, the generative adversarial imputation network is used to repair the missing samples, and the su-per-resolution technique is used to realize the super-resolution measurement of broadband os-cillation. In the oscillation modal analysis, the spectrum of the oscillation signal is extracted using the fast Fourier transform. To accurately locate the oscillation source, the branch potential energy is used as input to the data-driven model, such as LSTM and CNN. Finally, an oscillation source localization system is developed based on the digital twin workshop CloudPSS-XStudio, which can locate the oscillation source quickly and accurately.

Image fusion is the process of combining multiple input images from single or multiple imaging modalities into a fused image, which is expected to be more informative for human or machine perception as compared to any of the input images. In this paper, we propose a novel method based on deep learning for fusing infrared images and visible images, named the LBP-based proportional input generative adversarial network (LPGAN). In the image fusion task, the preservation of structural similarity and image gradient information is contradictory, and it is difficult for both to achieve good performance at the same time. To solve this problem, we innovatively introduce Local Binary Patterns (LBP) into Generative Adversarial Networks (GANs), which effectively utilize the texture features of the source images, so that the network has stronger feature extraction ability and anti-interference ability. In the feature extraction stage, we introduce a pseudo-siamese network for the generator to extract the detailed features and the contrast features. At the same time, considering the characteristic distribution of different modal images, we propose a 1:4 scale input mode. Extensive experiments on the publicly available TNO dataset and CVC14 dataset show that the proposed method achieves the state-of-the-art performance. We also test the universality of LPGAN through the fusion of RGB and infrared images on the RoadScene dataset. In addition, LPGAN is applied to multi-spectral remote sensing image fusion. Both qualitative and quantitative experiments demonstrate that our LPGAN can not only achieve good structural similarity, but also retain rich detailed information.

With the development of deep learning technology, various structures and research methods for super-resolution restoration of natural images and document images have been introduced. In particular, a number of recent studies have been conducted and developed in image restoration using generative adversarial network. Super-resolution restoration is ill-posed problem because of some complex restraints such as a lot of high-resolution images being restored for the same low-resolution image and also difficulty in restoring noises like edges, light smudging, and blurring. In this study, we utilized the spatially adaptive de-normalization (SPADE) structure for document image restoration to solve previous problems such as edge unclearness, hardness to catch features of texts, and the image color transition. Consequently, it can be confirmed that the edge of the character and the ambiguous stroke are restored more clearly when contrasting with the other previously suggested methods. Also, the proposed method’s PSNR and SSIM scores are geting 8% and 15% higher, respectively, compared to the previous methods.

Deep learning techniques, such as convolutional neural networks (CNN), generative adversarial networks (GAN), and graph neural networks (GNN), have over the past decade changed the ac-curacy of prediction in many diverse fields. In recent years, the application of deep learning tech-niques in computer vision tasks in pathology demonstrated extraordinary potential in assisting clinicians, automating diagnosis, and reducing costs for patients. Formerly unknown pathologi-cal evidence, such as morphological features related to specific biomarkers, copy number varia-tions, and other molecular features, were also able to be captured by deep learning models. In this paper, we review popular deep learning methods and some recent publications about their appli-cations in pathology.

The ‘inverse problem’ of mass spectrometric molecular identification (‘given a mass spectrum, calculate/predict the 2D structure of the molecule whence it came’) is largely unsolved, and is especially acute in metabolomics where many small molecules remain unidentified. This is largely because the number of experimentally available electrospray mass spectra of small molecules is quite limited. However, the forward problem (‘calculate a small molecule’s likely fragmentation and hence at least some of its mass spectrum from its structure alone’) is much more tractable, because the strengths of different chemical bonds are roughly known. This kind of molecular identification problem may be cast as a language translation problem in which the source language is a list of high-resolution mass spectral peaks and the ‘translation’ a representation (for instance in SMILES) of the molecule. It is thus suitable for attack using the deep neural networks known as transformers. We here present MassGenie, a method that uses a transformer-based deep neural network, trained on ~6 million chemical structures with augmented SMILES encoding and their paired molecular fragments as generated in silico, explicitly including the protonated molecular ion. This architecture (containing some 400 million elements) is used to predict the structure of a molecule from the various fragments that may be expected to be observed when some of its bonds are broken. Despite being given essentially no detailed nor explicit rules about molecular fragmentation methods, isotope patterns, rearrangements, neutral losses, and the like, MassGenie learns the effective properties of the mass spectral fragment and valency space, and can generate candidate molecular structures that are very close or identical to those of the ‘true’ molecules. We also use VAE-Sim, a previously published variational autoencoder, to generate candidate molecules that are ‘similar’ to the top hit. In addition to using the ‘top hits’ directly, we can produce a rank order of these by ‘round-tripping’ candidate molecules and comparing them with the true molecules, where known. As a proof of principle, we confine ourselves to positive electrospray mass spectra from molecules with a molecular mass of 500Da or lower, including those in the last CASMI challenge (for which the results are known), getting 49/93 (53%) precisely correct. The transformer method, applied here for the first time to mass spectral interpretation, works extremely effectively both for mass spectra generated in silico and on experimentally obtained mass spectra from pure compounds. It seems to act as a Las Vegas algorithm, in that it either gives the correct answer or simply states that it cannot find one. The ability to create and to ‘learn’ millions of fragmentation patterns in silico, and therefrom generate candidate structures (that do not have to be in existing libraries) directly, thus opens up entirely the field of de novo small molecule structure prediction from experimental mass spectra.

Data sets for medical images are generally imbalanced and limited in sample size because of high data collection costs, time-consuming annotations, and patient privacy concerns. The training of deep neural network classification models on these data sets to improve the generalization ability does not produce the desired results for classifying the medical condition accurately and often overfit the data on the majority of class samples. To address the issue, we propose a framework for improving the classification performance metrics of deep neural network classification models using transfer learning: pre-trained models, such as Xception, InceptionResNet, DenseNet along with the Generative Adversarial Network (GAN) – based data augmentation. Then, we trained the network by combining traditional data augmentation techniques, such as randomly flipping the image left to right and GAN-based data augmentation, and then fine-tuned the hyper-parameters of the transfer learning models, such as the learning rate, batch size, and the number of epochs. With these configurations, the Xception model outperformed all other pre-trained models achieving a test accuracy of 98.7%, the precision of 99%, recall of 99.3%, f1-score of 99.1%, receiver operating characteristic (ROC) - area under the curve (AUC) of 98.2%.

The question of molecular similarity is core in cheminformatics, and is usually assessed via a pairwise comparison based on vectors of properties or molecular fingerprints. We recently exploited variational autoencoders to embed 6M molecules in a chemical space, such that their (Euclidean) distance within the latent space so formed could be assessed within the framework of the entire molecular set. However, the standard objective function used did not seek to manipulate the latent space so as to cluster the molecules based on any perceived similarity. Using a set of some 160,000 molecules of biological relevance, we here bring together three modern elements of deep learning to create a novel and disentangled latent space, viz transformers, contrastive learning, and an embedded autoencoder. The effective dimensionality of the latent space was varied such that clear separation of individual types of molecules could be observed within individual dimensions of the latent space. The capacity of the network was such that many dimensions were not populated at all. As before, we assessed the utility of the representation by comparing clozapine with its near neighbours, and did the same for various antibiotics related to flucloxacillin. Transformers, especially when as here coupled with contrastive learning, effectively provide one-shot learning, and lead to a successful and disentangled representation of molecular latent spaces that at once uses the entire training set in their construction while allowing ‘similar’ molecules to cluster together in an effective and interpretable way.

The upsurge of Generative Adversarial Networks (GANs) in the previous five years has led to advancements in unsupervised data manipulation, sourced feature translation, and precise input-output synthesis through a competitive optimization of the discriminator and generator networks. More specifically, the recent rise of cycle-consistent GANs enables style transfers from a discrete source (input A) to target domain (input B) by preprocessing object features for a multi-discriminative adversarial network. Traditionally, cyclical adversarial networks have been exploited for unpaired image-to-image translation and domain adaptation by determining mapped relationships between an input A graphic and an input B graphic. However, this integral mechanism of domain adaptation can be applied to the complex acoustical features of human speech. Although well-established datasets, such as the 2018 Voice Conversion Challenge repository, paved way for female-male voice transformation, cycle-GANs have rarely been re-engineered for voices outside the datasets. More critically, cycle-GANs have massive potential to extract surface-level and hidden feature to distort an input A source into a texturally unrelated target voice. By preprocessing, compressing, and packaging unique acoustical voice properties, CycleGANs can learn to decompose speech signals and implement new translation models while preserving emotion, the intent of words, rhythm, and accents. Due to the potential of CycleGAN’s autoencoder in realistic unsupervised voice-voice conversion/feature adaptation, the researchers raise the ethical implications of controlling source input A to manipulate target voice B, particularly in cases of defamation and sabotage of target B’s words. This paper analyzes the potential of cycle-consistent GANs in deceptive voice-voice conversion by manipulating interview excerpts of political candidates.

The monitoring of fog density is of great importance in meteorology and its applications in environment, aviation and transportation. Nowadays, vision-based fog estimation from images taken with surveillance cameras has made a great supplementary contribution to the scarcely traditional meteorological fog observation. In this paper, we propose a new Random Forest (RF) approach for image-based fog estimation. In order to reduce the impact of data imbalance on recognition, the StyleGAN2-ADA (generative adversarial network with adaptive discriminator augmentation) algorithm is used to generate virtual images to expand the data of low proportions. Key image features related to fog are extracted, and an RF method, integrated with the hierarchical and k-medoids clustering, is deployed to estimate the fog density. The experiment conducted in Sichuan in February 2024 shows that the improved RF model has achieved an average accuracy of fog density observation to 93%, 6.4% higher than the RF model without data expansion, 3%-6% higher than the VGG16, the VGG19, the ResNet50, and the DenseNet169 with or without data expansion. What is more, the improved RF method exhibits a very good convergence as a cost-effective solution.

In the context of crowds innovation and the generative design driven by big language models, the personalized requirements has become the best engine for innovative design. The exploration of personalized requirements has become a key in significantly improving product innovation, concepts feasibility, and design interaction efficiency. To mine a large number of vague and unexpressed implicit requirements of personalized products, a domain knowledge graph-based method is proposed in this research. First, based on the classical theory of design science, the characteristics and categories of personalized implicit requirements are analyzed. Next, in order to improve the practicability and construction efficiency of the domain knowledge graph, a more informative ontology is constructed, and better performing NLP models are proposed. Then, a series of mining methods based on knowledge graph are proposed for the personalized implicit requirements of different categories. Finally, A platform was developed based on the technical solution proposed in this study, and an example verification was conducted in the field of electromechanical engineering. The efficiency improvement of the training model proposed in the research was analyzed, and the practicality of implicit requirement mining methods were discussed.

Facial recognition systems are widely used systems around the globe. Their applications in real world scenarios like law enforcement, surveillance, ID card identification, information security, access control and many more has made them very important in today’s world. These systems have shown promising results and excellent performance; however, these systems need tremendous number of facial images for training before they are implemented in the real-world scenarios. The problem arises when we face lack of training images due to which the performance of such systems decreases exponentially. The problem becomes even worse when there is only one image per subject for training our facial recognition system, the probe image might contain variation like illumination, expression and disguise which are hard to be accurately recognized. Generative models have achieved great success for handling the lack of training data by generating synthetic data that resembles the original data. We have created a new method by fine tuning the basic Conditional Generative Adversarial Network (CGAN) to generate synthetic facial images provided only single image per subject. These synthetic facial images contain six variations that are smile, anger, stare, illumination and disguise. We increased the size of the dataset and then applied a convolutional neural network for facial recognition, which improved the accuracy of our system from 0.76 to 0.99.

Social media has become more popular in the last few years. It has been used in public health development and healthcare settings to promote healthier lifestyles. Given its important role in today's culture, it is necessary to understand its current trends and future directions in public health. This review aims to describe and summarize how public health professionals have been using social media to improve population outcomes. Additionally, it discusses potential future directions on how social media can be used to improve population health. The findings showed how social media has been used as a tool for research, designing and implementing health campaigns, messaging, and health promotion. In addition, social media integration with Artificial Intelligence (AI) and Generative Pre-Trained transformers (GPT) can have a big impact and offer an innovative approach to tackle the problems and difficulties in health informatics. These innovations might improve patient participation, individualized assistance, and care experiences. The research showed how likely social media will keep growing and evolving and, if used effectively, have the potential to help close public health gaps across different cultures and improve population health.

In contemporary society, an increasing number of individuals are identified as the "Indoor Generation". Prolonged periods spent indoors can have detrimental effects on our overall well-being. Consequently, the examination of biophilic indoor environments and their impact on occupant well-being and comfort has emerged as a significant area of study. Besides the construction of new biophilic buildings, numerous existing building stocks retain substantial social, economic, and environmental value. The need for stock renovation is considerable, revitalizing existing structures rather than embarking on new constructions. Swift feedback is crucial during the initial stages of renovation design, fostering consensus among all stakeholders. A comprehensive understanding of proposed renovation plans contributes to improved indoor environment design. Initially, generative adversarial networks (GANs) produce interior renovation drawings based on user preferences and designer styles. Subsequently, the implementation of Mixed Reality (MR) and Diminished Reality (DR) provides users with an immersive perspective, transforming 2D drawings into interactive MR experiences, and facilitating the presentation of interior renovation proposals. This paper outlines the development of a real-time system for architectural designers that seamlessly integrates MR, DR, and GANs results, with the aim of enhancing feedback efficiency during the renovation design process, enabling stakeholders to evaluate and understand renovation proposals more comprehensively. Furthermore, we incorporate several full-reference image quality assessment (FR-IQA) methods to evaluate the quality of the GANs-generated images. The evaluation results indicate that the majority of images fall within the moderate range of quality.

The disruptive application of ChatGPT (GPT-3.5, GPT-4) to a variety of domains has become a topic of much discussion in the scientific community and society at large. Large Language Models (LLMs), e.g., BERT, Bard, Generative Pre-trained Transformers (GPTs), LLaMA, etc., have the ability to take instructions, or prompts, from users and generate answers and solutions based on very large volumes of text-based training data. This paper assesses the impact and potential impact of ChatGPT on the field of digital forensics, specifically looking at its latest pre-trained LLM, GPT-4. A series of experiments are conducted to assess its capability across several digital forensic use cases including artefact understanding, evidence searching, code generation, anomaly detection, incident response, and education. Across these topics, its strengths and risks are outlined and a number of general conclusions are drawn. Overall this paper concludes that while there are some potential low-risk applications of ChatGPT within digital forensics, many are either unsuitable at present, since the evidence would need to be uploaded to the service, or they require sufficient knowledge of the topic being asked of the tool to identify incorrect assumptions, inaccuracies, and mistakes. However, to an appropriately knowledgeable user, it could act as a useful supporting tool in some circumstances.

This thesis presents the effects of work done on a software project for generative models and spreadsheets, allowing for a quick creation of the conceptual model of the aircraft. The subject of the work is a response to the current trends and needs prevailing in the field of computer design engineering CAD and aviation. In the initial chapters, theoretical issues related to the work being carried out were introduced and the methodology of creating software for construction and verification of the structure of aircraft along with the needs of interchange between databases of generative models was presented. In the next stages, the concepts and selected solutions for the user interface supporting the knowledge base were presented along with a set of procedures for its operation. Furthermore, the method of database integration with the methods of determining design features for the developed generative models and with the Siemens NX system. Furthermore, problems encountered in software development, as well as solution examples for model application are specified. The results obtained and the models generated on their basis were subjected to a strength analysis using Autodesk Inventor software and analysed in terms of meeting the initial assumptions. In the end, conclusions and observations resulting from the effects of the work presented in the project were formulated.

The QBIT theory is an attempt toward solving the problem of consciousness based on empirical evidence provided by various scientific disciplines including quantum mechanics, biology, information theory, and thermodynamics. This theory formulates the problem of consciousness in the following four questions: (1) What is the nature of qualia? (2) How are qualia generated? (3) Why are qualia subjective? (4) Why does a quale have a particular quality or meaning?In sum, the QBIT theory proposes that (1) when a pack of quantum information is compressed beyond a certain threshold, a quale is generated; (2) a quale is a superdense pack of maximally entangled qubits in a pure state; (3) when information-theoretic certainty of a system about an external stimulus exceeds a particular level, the system becomes conscious of that stimulus; (4) subjectivity of consciousness is due to the fact that maximally entangled pure states are private and unshareable.

Extreme weather events, such as heat waves and heavy precipitation, are becoming increasingly frequent due to climate change, necessitating the development of effective early warning systems (EWS) to mitigate their impacts. This study introduces an advanced EWS designed specifically for Italy, which integrates the FourCastNet global forecasting model with Elevation-integrated TEmperature and Precipitation SRGAN downscaling (E-TEPS) to enhance the spatial resolution and accuracy of climate predictions. Building on previous work that demonstrated E-TEPS's effectiveness in downscaling temperature and total precipitation variables, this research applies the integrated system to two severe weather events in Central Italy: the Emilia-Romagna floods of 2023 and the Marche floods of 2022. The system's performance was evaluated using key metrics, including Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Pearson Correlation, by comparing downscaled outputs from FourCastNet and ERA5 datasets against high-resolution ground truth data from the Euro-Mediterranean Center on Climate Change (CMCC) dataset. The results indicate that the integrated system offers improved predictive accuracy, particularly in capturing critical climate variables, with the entire EWS delivering high-resolution forecasts and final outputs in under one minute. Although potential limitations were identified due to biases in the underlying datasets, which could affect forecast reliability in regions with complex topography or extreme weather conditions, this research highlights the potential of combining machine learning models with downscaling techniques to enhance EWS precision, providing valuable insights for future climate forecasting and disaster preparedness strategies.

Groundnut is an oil seed cash crop, that can be consumed directly by humans and animals. It is also used as an intelligence in the industrial production of butter and other products. However, Groundnut seeds can become infected with fungi, viruses, pests, physical damage, and high heat, which can negatively impact crop yield, quality, and economic value, and pose health hazards. To address these challenges, computer vision technology detects and classifies defects. This study introduced an ensemble deep learning defected classification model that combines VGG16 and InceptionV3 using seed images. To enhance model performance, collected images are preprocessed using novel techniques tailored to different image types. Preprocessing techniques are chosen based on image-quality evaluation metrics. Watershed segmentation is applied followed by detecting the region of interest in the image using YOLOv3. The image dataset is augmented and balanced using a Generative Adversarial Network (GAN). The model development involves a combination of classical and deep-based features, comparing features extracted with (HOG and GLCM) to those extracted with InceptionV3 and VGG16. The ensemble model achieves an accuracy of 96.25% with a split ratio of 10% for testing, 10% for validation, and 80% for training set. This research benefits farmers looking to improve yield, consumers of groundnut products, and researchers studying seed defects. The work provides valid insights for future researchers regarding dataset creation, augmentation, methods, and feature selection for modeling. However, the model experienced misclassifications due to the similar appearance of sample images in different classes. Future researchers should address these challenges and consider factors such as oil content and defect grading.

Glaucoma, a global leading cause of blindness, necessitates accurate segmentation of the optic disc (OD) and optic cup (OC) for effective screening. However, existing segmentation methods based on convolutional neural networks (CNNs) suffer from high computational complexity and long inference times. In this paper, we propose an end-to-end segmentation method for the OD and OC using a lightweight MobileNetv3 as the feature extraction network. Our approach incorporates boundary branches and an adversarial learning network to achieve multi-label segmentation of the OD and OC. We evaluate the proposed method on three publicly available datasets: Drishti-GS, RIM-ONE-r3, and REFUGE. The experimental results demonstrate segmentation accuracies of 0.974/0.900, 0.966/0.875, and 0.962/0.880 for the OD and OC on the respective datasets, while significantly reducing inference time.

Hyperspectral small target detection (HSTD) is a promising pixel-level detection task. However, due to the low contrast and imbalance number between the target and background spatially and high dimensions spectrally, it is a challenging one. To address these issues, this work proposes a representation learning-based graph and generative network for hyperspectral small target detection. The model builds a fusion network through frequency representation for HSTD, where the novel architecture incorporates irregular topological data and the spatial-spectral feature to improve its representation ability. Firstly, a graph convolutional network (GCN) module better models the non-local topological relationship between samples to represent the hyperspectral scene’s underlying data structure. The mini-batch-training pattern of the GCN decreases the high computational cost of building an adjacency matrix for high-dimensional data sets. In parallel, the generative model enhances the differentiation reconstruction and the deep feature representation ability with respect to the target spectral signature. Finally, a fusion module compensates for the extracted different types of HS features and integrates their complementary merits for hyperspectral data interpretation while increasing detection and background suppression capabilities. Experiments on different hyperspectral data sets demonstrate the advantages of the proposed architecture.

Emerging semiconductor devices often lack accurate analytical models. The same is usually true of any devices working under extreme conditions like cryogenic temperatures. The usual workaround involves the use of approximation models, usually based on lookup tables or neural networks individually fitted to measurement data. In the case of experimental devices or ones working under extreme conditions, the number of units available for measurement is limited. As a result, the number of approximation-model instances is too small to enable a statistical simulation of even middle-sized circuits, which is a necessary step in integrated-circuit design since it provides the realistic picture of the circuit’s behavior in the presence of manufacturing process variations. Approximation models using structure parameters as inputs do exist in the literature, but are only useful if the end user knows the statistical distributions of those parameters, which is not usually the case. We propose a technique based on generative machine learning, namely the variational autoencoder, that uses only a small sample of devices to capture the essential features of their I–V curves under process variations and subsequently generates an arbitrary number of similarly disturbed curves. The model trained on as few as 20 instances per device type is shown to precisely reproduce the distributions of period and power consumption of a ring oscillator.

With the popularity of solar energy in the electricity market, demand arises for data such as precise locations of solar panels for efficient energy planning, management, and distribution. However, this data is not easily accessible and in some cases, information such as precise locations does not exist. Furthermore, existing data sets for training semantic segmentation models of PV installations are limited, and their annotation is time-consuming and labor-intensive. Therefore, for additional remote sensing (RS) data creation, the pix2pix generative adversarial network (GAN) is utilized, enriching the original resampled training data of varying GSDs without compromising its integrity. Experiments done with the DeepLabV3 model, ResNet-50 backbone, and pix2pix GAN architecture were conducted to find the optimal configuration for an accurate RS imagery segmentation model. The result is a fine-tuned solar panel semantic segmentation model, trained using transfer learning and utilizing an optimal amount – 60% of generated RS imagery for additional training data, increasing model accuracy. The findings demonstrate the benefits of using GAN-generated images as additional training data, increasing the size of small data sets, and improving the capabilities of the segmentation model for solar panel detection in RS images.

To realize fast and effective synthetic aperture radar (SAR) deception jamming, a high-quality SAR deception jamming template library can be generated by performing sample augmentation on SAR deception jamming templates. However, current sample augmentation schemes of SAR deception jamming templates face certain problems. First, the authenticity of templates is low due to the lack of speckle noise. Second, the generated templates have low similarity to the target and shadow areas of the input templates. To solve these problems, this study proposes a sample augmentation scheme based on generative adversarial networks, which can generate a high-quality library of SAR deception jamming templates with shadows. The proposed scheme solves the two aforementioned problems from the following aspects. First, the influence of the speckle noise is considered in the network to avoid the problem of reduced authenticity in generated images. Second, a channel attention mechanism module is used to improve the network's learning ability of shadow features, which improves the similarity between the generated template and the shadow area in the input template. Finally, the proposed scheme and the SinGAN scheme are compared regarding the equivalent numbers of looks and the structural similarity between the target and shadow in the sample augmentation results. The comparison results demonstrate that, compared to the templates generated by the SinGAN scheme, those generated by the proposed scheme have targets and shadow features similar to those of the original image, and can incorporate speckle noise characteristics, resulting in higher authenticity, which helps to achieve fast and effective SAR deception jamming.

The use of artificial intelligence in natural language processing (NLP) has significantly contributed to the advancement of natural language applications such as sentimental analysis, topic modeling, text classification, chatbots, and spam filtering. With a large amount of text generated each day from different sources such as webpages, blogs, emails, social media, and articles, one of the most common tasks in natural language processing is the classification of a text corpus. This is important in many institutions for planning, decision-making, and archives of their projects. Many algorithms exist to automate text classification operations but the most intriguing of them is that which also learns these operations automatically. In this study, we present a new model to infer and learn from data using probabilistic logic and apply it to text classification. This model, called GenCo, is a multi-input single-output (MISO) learning model that uses a collaboration of partial classifications to generate the desired output. It provides a heterogeneity measure to explain its classification results and enables the reduction of the curse of dimensionality in text classification. The classification results are compared with those of conventional text classification models, and it shows that our proposed model has a higher classification performance than conventional models.

In recent years, artificial intelligence technology has widely influenced the field of design, bringing new ideas to efficiently and systematically solve urban renewal design problems. The purpose of this study is to create a stylized generation technology for building facade decoration in historic districts, which will aid in the design and control of district style and form. The goal is to use the technical advantages of conditional generative adversarial network (CGAN) in image generation and style transfer to create a method for independently designing a specific facade decoration style by interpreting image data of historical district facades. The research in this paper is based on the historical district of Putian in Fujian Province, through an experiment of image data acquisition, image processing and screening, model training, image generation, and style matching of the target area. The research found that: (1) CGAN technology can better identify and generate the decorative style of historical districts. It can realize the overall or partial scheme design of the facade; (2) in terms of adaptability, this method can provide a better scheme reference for historical district reconstruction, facade renovation, and renovation design projects. Especially for districts with obvious decorative styles, the visualization effect is better. In addition, it also has certain reference significance for the determination and design of the facade decoration style of a specific historical building; (3) This method can better learn the internal laws of the complex district style and form so as to generate a new design with a clear decoration style attribute. It can be extended to other fields of historical heritage protection to enhance practitioners' stylized control of the heritage environment and improve the efficiency and ability of professional design.

Machine Learning (ML) methods have become important to enhance the performance of decision-support predictive models. However, class imbalance is one of the main challenges for developing ML models, because it limits the generalization of these models, and biases the learning algorithms. In this paper, we consider oversampling methods for generating synthetic categorical clinical data aiming to improve the predictive performance in ML models, and the identification of risk factors for cardiovascular diseases (CVDs). We performed a comparative study of several categorical synthetic data generation methods, including Generative Adversarial Networks (GANs). Then, we assessed the impact of combining oversampling strategies and linear and nonlinear supervised ML methods. Lastly, we conducted a post-hoc model interpretability based on the importance of the risk factors. Experimental results show the potential of GAN-based models for generating high-quality categorical synthetic data, yielding probability mass functions that are highly close to real data, maintaining relevant insights, and contributing to increase the predictive performance. The GAN-based model and a linear classifier outperforms other oversampling techniques, improving 2\% the area under the curve. These results demonstrate the capability of synthetic data to help both in determining risk factors and building models for CVD prediction.

Deep learning (DL), a branch of machine learning (ML) and artificial intelligence (AI) is nowadays considered as a core technology of today's Fourth Industrial Revolution (4IR or Industry 4.0). Due to its learning capabilities from data, DL technology originated from artificial neural network (ANN), has become a hot topic in the context of computing, and is widely applied in various application areas like healthcare, visual recognition, cybersecurity, and many more. However, building an appropriate DL model is a challenging task, due to the dynamic nature and variations in real-world problems and data. Moreover, the lack of core understanding turns DL methods into black-box machines that hamper development at the standard level. This article presents a structured and comprehensive view on DL techniques including a taxonomy considering various types of real-world tasks like supervised or unsupervised. In our taxonomy, we take into account deep networks for supervised or discriminative learning, unsupervised or generative learning as well as hybrid learning and relevant others. We also summarize real-world application areas where deep learning techniques can be used. Finally, we point out ten potential aspects for future generation DL modeling with research directions. Overall, this article aims to draw a big picture on DL modeling that can be used as a reference guide for both academia and industry professionals.

Synesthesia is a psychological phenomenon where sensory signals become mixed. Input to one sensory modality produces an experience in a second, unstimulated modality. In “grapheme-color synesthesia”, viewed letters and numbers evoke mental imagery of colors. The study of this condition has implications for increasing our understanding of brain architecture and function, language, memory and semantics, and the nature of consciousness. In this work, we propose a novel application deep learning to model perception in grapheme-color synesthesia. Achromatic letter images, taken from database of handwritten characters, are used to induce synesthesia. Results show the model learns to accurately create a colored version of the inducing stimulus, according to a statistical distribution from experiments on a sample population of grapheme-color synesthetes. The spontaneous, creative mental imagery characteristic of the synesthetic perceptual experience is reproduced by the model. A model of synesthesia that generates testable predictions on brain activity and behavior is needed to complement large scale data collection efforts in neuroscience, especially when articulating simple descriptions of cause (stimulus) and effect (behavior). The research and modeling approach reported here begins to address this need.

Traditional urban design often overlooks the synchronisation of human and ecological connectivities, typically favouring corridors for ecological continuity. Our study challenges this convention by introducing a computational design approach, Meta-Connectivity, leveraging the deep generative models performing cross-domain translation to integrate human-wildlife landscape connectivity in urban morphology amidst the planetary urbanisation. Utilising chained Pix2Pix models, our research illustrates a novel meta-connectivity design reasoning framework, combining landscape connectivity modelling with conditional reasoning based on deep generative models. This framework enables the adjustment of both human and wildlife landscape connectivities based on their correlative patterns in one single design process, guiding the re-materialisation of urban landscapes without the need for explicit prior ecological or urban data. Our empirical study in East London demonstrated the framework's efficacy in suggesting wildlife connectivity adjustments based on human connectivity metrics. The results demonstrate the feasibility of creating an innovative urban form in which the land cover guided by the connectivity gradients replaces the corridors based on simple geometries. This research thus presents a methodology shift in urban design, proposing a symbiotic approach to integrating disparate yet interrelated landscape connectivities within urban contexts.

In recent years, the rapid development of computer-aided planning and design technology has provided a new perspective for the study of complex problems such as the generation of architectural complex forms. This study focuses on how to apply parameterization technology to analyze and articulate traditional spatial form composition rules, aiming to minimize reliance on subjective human judgment in the protection and renewal design of the historical style of traditional villages. It aims to establish digital generative design tools to address the challenges of accurately inheriting and innovatively utilizing historical and cultural information in traditional settlements. It introduces how to rely on parameterization technology to analyze the spatial form composition rules, parameter extraction rules, and spatial reconstruction rules of traditional villages, facilitating the complete process from spatial features to parameterization rules, and then to the application of computational methods to deduce spatial features. It also includes case studies demonstrating the application of parameterization technology tools for village protection and explores the role of generative design tools in preserving the spatial style of these settlements.

Generative models have recently gained popularity in Remote Sensing, offering substantial benefits for interpreting and utilizing satellite imagery across diverse applications such as climate monitoring, urban planning, and wildfire detection. These models are particularly adept at addressing the challenges posed by satellite images, which often exhibit domain variability due to seasonal changes, sensor characteristics, and especially variations in spectral bands. Such variability can significantly impact model performance across various tasks. In response to these challenges, our work introduces a novel approach that harnesses the capabilities of Generative Adversarial Networks (GANs), augmented with Contrastive Learning, to generate target domain images that account for multi-spectral band variations effectively. By maximizing mutual information between corresponding patches and leveraging the power of GANs, our model aims to generate realistic-looking images across different multi-spectral domains. We present a comparative analysis of our model against other well-established generative models, demonstrating its efficacy in generating high-quality satellite images while effectively managing domain variations inherent to multi-spectral diversity.

Research on image-inpainting tasks has mainly focused on enhancing performance by augmenting various stages and modules. However, this trend does not consider the increase in the number of model parameters and operational memory, which increases the burden on computational resources. To solve this problem, we propose a Parametric Efficient Image InPainting Network (PEIPNet) for efficient and effective image-inpainting. Unlike other state-of-the-art methods, the proposed model has a one-stage inpainting framework in which depthwise and pointwise convolutions are adopted to reduce the number of parameters and computational costs. To generate semantically appealing results, we selected three unique components: spatially-adaptive denormalization (SPADE), dense dilated convolution module (DDCM), and efficient self-attention (ESA). The SPADE was adopted to conditionally normalize activations according to the mask to distinguish between damaged and undamaged regions. The DDCM was employed at every scale to overcome the gradient-vanishing obstacle and gradually fill-in pixels by capturing global information along the feature maps. The ESA was utilized to obtain clues from unmasked areas by extracting long-range information. In terms of efficiency, our model has the lowest operational memory compared with other state-of-the-art methods. Both qualitative and quantitative experiments demonstrate the generalized inpainting of our method on three public datasets: Paris StreetView, CelebA, and Places2.

Background: ChatGPT is becoming a new reality. Where do we go from here? Objective: to show how we can distinguish ChatGPT-generated publications from counterparts produced by scientist. Methods:By means of a new algorithm, called xFakeBibs, we show the significant difference between ChatGPT-generated fake publications and real publications. Specifically, we triggered ChatGPT to generate 100 publications that were related to Alzheimer’s disease and comorbidity. Using TF-IDF, using the real publications, we constructed a training network of bigrams comprised of 100 publications. Using 10-folds of 100 publications each, we also 10 calibrating networks to derive lower/upper bounds for classifying articles as real or fake. The final step was to test xFakeBibs against each of the ChatGPT-generated articles and predict its class. The algorithm successfully assigned the POSITIVE label for real ones and NEGATIVE for fake ones. Results: When comparing the bigrams of the training set against all the other 10 calibrating folds, we found that the similarities fluctuated between (19%-21%). On the other hand, the mere bigram similarity from the ChatGPT was only (8%). Additionally, when testing how the various bigrams generated from the calibrating 10-folds against ChatGPT we found that all 10 calibrating folds contributed (51%-70%) of new bigrams, while ChatGPT contributed only 23%, which is less than 50% of any of the other 10 calibrating folds. The final classification results using the xFakeBibs set a lower/upper bound of (21.96-24.93) number of new edges to the training mode without contributing new nodes. Using this calibration range, the algorithm predicted 98 of the 100 publications as fake, while 2 articles failed the test and were classified as real publications. Conclusions: This work provided clear evidence of how to distinguish, in bulk ChatGPT-generated fake publications from real publications. Also, we also introduced an algorithmic approach that detected fake articles with a high degree of accuracy. However, it remains challenging to detect all fake records. ChatGPT may seem to be a useful tool, but it certainly presents a threat to our authentic knowledge and real science. This work is indeed a step in the right direction to counter fake science and misinformation.

In actuarial practice, the modeling of total losses tied to a certain policy is a non-trivial task. Traditional parametric models to predict total losses have limitations due to complex distributional features such as extreme skewness, zero inflation, multi-modality, etc., and the lack of explicit solutions for log-normal convolution. In the recent literature, the application of the Dirichlet process mixture for insurance loss has been proposed to eliminate the risk of model misspecification biases; however, the effect of covariates as well as missing covariates in the modeling framework is rarely studied. In this article, we propose novel connections among covariate-dependent Dirichlet process mixture, log-normal convolution, and missing covariate imputation. Assuming an individual loss is log-normally distributed, we develop a log skew-normal Dirichlet process to approximate the log-normal sum. As a generative approach, our framework models the joint of outcome and covariates, which allows to impute missing covariates under the assumption of missingness at random. The performance is assessed by applying our model to several insurance datasets, and the empirical results demonstrate the benefit of our model compared to the existing actuarial models such as the Tweedie-based generalized linear model, generalized additive model, or multivariate adaptive regression spline.

Augment reality (AR) is crucial for immersive human-computer interaction (HCI) and vision of artificial intelligence (AI). Labeled data drove object recognition in AR. However, manual annotating data is expensive and labor-intensive, and furthermore, scanty labeled data limits the application of AR. Aiming at solving the problem of insufficient training data in AR object recognition, an automated vision data synthesis method called BAGAN is proposed in this paper based on the 3D modeling and GAN algorithm. Our approach has been validated to have better performance than other methods through image recognition task on natural image database ObjectNet3D. This study can shorten the algorithm development time of AR and expand the application scope of AR, which is of great significance to immersive interactive systems.

The integration of deep learning models into financial risk prediction and analysis has significantly transformed traditional approaches. While conventional quantitative methods often rely on simplistic metrics like maximum drawdown, the advent of deep learning necessitates a more nuanced evaluation, emphasizing the model's generalization ability, especially during market crises such as stock market crashes. This paper explores the critical aspects of evaluating deep learning models' risk control capabilities in finance, underscoring the importance of understanding both statistical metrics and generalization abilities, particularly in adverse market conditions. By examining deep learning models' performance in scenarios like stock market crashes and highlighting the significance of cross-validation techniques, this study aims to offer practitioners insights into constructing robust risk management systems. It advocates for a comprehensive approach integrating quantitative analysis with macroeconomic factors to enhance financial risk prediction and analysis in volatile markets. The experimentation reveals that different deep generative models excel in various aspects of financial time series analysis, with generative adversarial networks (GANs) demonstrating superior performance in predicting Value at Risk (VaR) and variational autoencoders (VAEs) excelling in return rate prediction. Moreover, integrating multiple models further enhances predictive performance, leveraging the strengths of each model to compensate for individual weaknesses. Overall, this paper underscores the potential and significance of deep generative models in financial time series analysis, offering a roadmap for improved risk management and decision-making in financial markets.

Business-to-business (B2B) manufacturing companies are increasingly confronted with transformative trends such as sustainability, digitalization, and servitization. These trends are changing how product portfolios are developed, and value contributions are assessed, and therefore have disruptive potential. Dealing with these disruptive factors in Product Portfolio Management (PPM) is a largely unexplored topic. This study presents an empirical-qualitative exploration that contributes significantly to the field. The aim is to clarify the extent to which disruptive factors influence the evaluation and shaping of the product portfolio in B2B manufacturing companies. The Gioia method was used to evaluate 21 semi-structured interviews with experts from leading B2B manufacturing companies. Eight overarching challenges in PPM resulting from the disruptive factors were identified. Based on the eight overarching challenges and their associated causal relationships, two aggregated dimensions of action were derived: (1) increasing speed and flexibility by using generative artificial intelligence (AI) in a defined PPM process and (2) adjusting the product portfolio evaluation to consider various strategic drivers. These two dimensions of action call for future research to overcome the disruptive factors in PPM.

Structural design optimization (SDO) plays a pivotal role in enhancing various aspects of construction projects, including design quality, cost-efficiency, safety, and structural reliability. Recent endeavors in academia and industry have sought to harness the potential of Building Information Modeling (BIM) and optimization algorithms to optimize SDO and improve design outcomes. This review paper aims to synthesize these efforts, shedding light on how SDO contributes to project coordination. Furthermore, the integration of sustainability considerations and the application of innovative technologies and optimization algorithms in SDO necessitate more interactive early-stage collaboration among project stakeholders. This study offers a comprehensive exploration of contemporary research in integrated SDO employing BIM and optimization algorithms. It commences with an exploratory investigation, employing both qualitative and quantitative analysis techniques following the PRISMA systematic review methodology. Subsequently, an open-ended opinion survey was conducted among construction industry professionals in Europe. This survey yields valuable insights into the coordination challenges and potential solutions arising from technological shifts and interoperability concerns associated with widespread SDO implementation. These preliminary steps of systematic review and industry survey furnish a robust knowledge foundation, enabling the proposal of an intelligent framework for automating early-stage sustainable structural design optimization (ESSDO) within the construction sector. The framework ESSDO addresses the challenges of fragmented collaboration between architects and structural engineers. This proposed framework seamlessly integrates with the BIM platform, i.e., Autodesk Revit for architects. It extracts crucial architectural data and transfers it to the structural design and analysis platform, i.e., Autodesk Robot Structural Analysis (RSA), for structural engineers via the visual programming tool Dynamo. Once the optimization occurs, optimal outcomes are visualized within BIM environments. This visualization elevates interactive collaborations between architects and engineers, facilitating automation throughout the workflow and smoother information exchange.

OpenAI has managed to turn 100 million heads in two months towards their new Language model tool (LM) ChatGPT. The third generation Generative Pretrained Transformer (GPT-3) has the capacity to tackle simple to complex and sophisticated problems, while providing reasoning behind its generated answers. ChatGPT can be used to increase productivity and improve efficiency and face challenging problems. Programming mobile applications is a challenging task that requires professional software engineers, skills and abilities to be developed. The following paper takes a case study approach to assess how novice app developers can use ChatGPT to generate Java scripts that will be used in Android studio to create a functional application. The results after, many iterations and ongoing conversations with ChatGPT managed to create an application for the anticipated function. Important insights have been drawn from the case study that could set the ground for any novice user seeking to create applications using Java scripting and Android Studio.

Convolutional Neural Networks combined with autonomous drones are increasingly seen as enablers of partially automating the aircraft maintenance visual inspection process. Such an innovative concept can have a significant impact on aircraft operations. Through supporting aircraft maintenance engineers detect and classify a wide range of defects, the time spent on inspection can significantly be reduced. Examples of defects that can be automatically detected include aircraft dents, paint defects, cracks and holes, and lightning strike damage. Additionally, this concept could also increase the accuracy of damage detection and reduce the number of aircraft inspection incidents related to human factors like fatigue and time pressure. In our previous work, we have applied a recent Convolutional Neural Network architecture known by MASK R-CNN to detect aircraft dents. MASK-RCNN was chosen because it enables the detection of multiple objects in an image while simultaneously generating a segmentation mask for each instance. The previously obtained F1 and F2 scores were 62.67% and 59.35% respectively. This paper extends the previous work by applying different techniques to improve and evaluate prediction performance experimentally. The approaches uses include (1) Balancing the original dataset by adding images without dents; (2) Increasing data homogeneity by focusing on wing images only; (3) Exploring the potential of three augmentation techniques in improving model performance namely flipping, rotating, and blurring; and (4) using a pre-classifier in combination with MASK R-CNN. The results show that a hybrid approache combining MASK R-CNN and augmentation techniques leads to an improved performance with an F1 score of (67.50%) and F2 score of (66.37%)

Despite numerous studies in statistical downscaling methodology, there remains a lack of methods that can downscale from precipitation modeled in global climate models to regional level high resolution gridded precipitation. This paper reports a novel downscaling method using a Generative Adversarial Network (GAN), CliGAN, which can downscale large-scale annual maximum precipitation given by simulation of multiple atmosphere-ocean global climate models (AOGCM) from Coupled Model Inter-comparison Project 6 (CMIP6) to regional-level gridded annual maximum precipitation data. This framework utilizes a convolution encoder-dense decoder network to create a generative network and a similar network to create a critic network. The model is trained using an adversarial training approach. The critic uses the Wasserstein distance loss function and the generator is trained using a combination of adversarial loss Wasserstein distance, structural loss with the multi-scale structural similarity index (MSSIM), and content loss with the Nash-Sutcliff Model Efficiency (NS). The MSSIM index allowed us to gain insight into the model’s regional characteristics and shows that relying exclusively on point-based error functions, widely used in statistical downscaling, may not be enough to reliably simulate regional precipitation characteristics. Further use of structural loss functions within CNN-based downscaling methods may lead to higher quality downscaled climate model products.

This study attempts to classify compound words on the basis of Cognitive Linguistics and compares their usage trends using Computational Linguistics. Using Noam Chomsky’s concept of deep and surface structures of a sentence, Lees treated compound words, not as separate units but as a kind of embedded sentences and hinted for possible presence of deep and surface structures in compound words, which this study tries to investigate. Then on the basis of the Idealized Cognitive Model proposed by Lakoff and Fauconnier, compound words have been classified into transparent, opaque and counterintuitive compound words. Using Google Books Corpus, this study also compares their usage trends. This is done using usage frequency, defined in this work, which is analogous to productivity for affixed words calculated by G.E.Booij. Each class of compound word formed on the basis of ICM is found to have different usage frequency. The possible reasons for this are discussed.

Humans have a unique cognitive ability known as relational abstraction, which allows us to identify logical rules and patterns beyond basic perceptual characteristics. This ability represents a key difference between how humans and other animals learn and interact with the world. With current large language models rivaling human intelligence in many regards, we investigated whether relational abstraction was an emergent chat completion capacity in various models. We find that despite their impressive language processing skills, all tested language models failed the relational match-to-sample (RMTS) test, a benchmark for assessing relational abstraction. These results challenge the assumption that advanced language skills inherently confer the capacity for complex relational reasoning. The paper highlights the need for a broader evaluation of AI cognitive abilities, emphasizing that language proficiency alone may not be indicative of certain higher-order cognitive processes thought to be supported by language.

The migration from MATLAB to Python (M-to-PY) has gained significant traction in recent computational research. While MATLAB has long served as a linchpin in myriad scientific endeavors, there's an emerging trend to rejuvenate these projects using Python's extensive AI tools and libraries. This study presents a semi-automated process for M-to-PY conversion, using a detailed case study of an image skeletonization project comprising fifteen MATLAB files and a 1404-image dataset. Skeletonization is foundational for ongoing 3D motion detection research using AI transformers, predominantly developed in Python. The utilization of ChatGPT-4, acting as an AI co-programmer, is pivotal in this conversion. By leveraging the public OpenAI API, we developed an M-to-PY converter prototype, evaluated its efficacy using test cases from the Bard bot, and subsequently employed the converted code in an AI application. The dual contributions encompass a well-tested M-to-PY converter and a Skeleton App capable of sketching and skeletonizing any given image, enriching the AI toolset. This study accentuates how AI resources, like ChatGPT-4, can simplify code transitions, opening doors for innovative AI implementations using primarily MATLAB-coded scientific research.

Computed tomography (CT) offers detailed insights into the internal anatomy of patients, partic-ularly for spinal vertebrae examination. However, CT scans are associated with higher radiation exposure and cost compared to conventional X-ray imaging. In this study, we applied a Genera-tive Adversarial Network (GAN) framework to reconstruct 3D spinal vertebrae structures from synthetic biplanar X-ray images, specifically focusing on anterior and lateral views. The synthetic X-ray images were generated using the DRRGenerator module in 3D Slicer, by incorporating segmentations of spinal vertebrae in CT scans for the region of interest focussing. The approach leverages a novel feature fusion technique based on X2CT-GAN to combine information from both views and employs a combination of Mean Squared Error (MSE) loss and adversarial loss to train the generator, resulting in high-quality synthetic 3D spinal vertebrae CTs. A total of n=440 CT data were processed. We evaluated the performance of our model using multiple metrics, in-cluding Mean Absolute Error (MAE) (for each slice of the 3D volume [MAE0] and for the entire 3D volume [MAE]), Cosine Similarity, Peak Signal-to-Noise Ratio (PSNR), 3D Peak Sig-nal-to-Noise Ratio (PSNR-3D), and Structural Similarity Index (SSIM). The average PSNR was 28.394 dB, PSNR-3D was 27.432, SSIM was 0.468, cosine similarity was 0.484, MAE0 was 0.034, and MAE was 85.359. The results demonstrated the effectiveness of the approach in reconstruct-ing 3D spinal vertebrae structures from biplanar X-rays, although some limitations in accurately capturing the fine bone structures and maintaining the precise morphology of the vertebrae were present. This technique has the potential to enhance the diagnostic capabilities of low-cost X-ray machines while reducing radiation exposure and cost associated with CT scans, paving the way for future applications in spinal imaging and diagnosis.

Drone and aerial remote sensing images are widely used, but their imaging environment is complex and prone to image blurring. Existing CNN deblurring algorithms usually use multi-scale fusion to extract features in order to make full use of aerial remote sensing blurred image information, but images with different degrees of blurring use the same weights, leading to increasing errors in the feature fusion process layer by layer. Based on the physical properties of image blurring, this paper proposes an adaptive multi-scale fusion blind deblurred generative adversarial network (AMD-GAN), which innovatively applies the degree of image blurring to guide the adjustment of the weights of multi-scale fusion, effectively suppressing the errors in the multi-scale fusion process and enhancing the interpretability of the feature layer. The research work in this paper reveals the necessity and effectiveness of a priori information on image blurring levels in image deblurring tasks. By studying and exploring the image blurring levels, the network model focuses more on the basic physical features of image blurring. Meanwhile, this paper proposes an image blurring degree description model, which can effectively represent the blurring degree of aerial remote sensing images. The comparison experiments show that the algorithm in this paper can effectively recover images with different degrees of blur, obtain high-quality images with clear texture details, outperform the comparison algorithm in both qualitative and quantitative evaluation, and can effectively improve the object detection performance of aerial remote sensing blurred images. Moreover, the average PSNR of this paper's algorithm tested on the publicly available dataset RealBlur-R reached 41.02dB, surpassing the latest SOTA algorithm.

Despite the importance of few-shot learning, the lack of labeled training data in the real world, makes it extremely challenging for existing machine learning methods as this limited data set does not represent the data variance well. In this research, we suggest employing a generative approach using variational autoencoders (VAEs), which can be used specifically to optimize few-shot learning tasks by generating new samples with more intra-class variations. The purpose of our research is to increase the size of the training data set using various methods to improve the accuracy and robustness of the few-shot face recognition. Specifically, we employ the VAE generator to increase the size of the training data set, including the basic and the novel sets while utilizing transfer learning as the backend. Based on extensive experimental research, we analyze various data augmentation methods to observe how each method affects the accuracy of face recognition. We conclude that the face generation method we proposed can effectively improve the recognition accuracy rate to 96.47% using both the base and the novel sets.

Social Anxiety Disorder (SAD) is a common mental health condition characterized by an intense fear of social situations, leading to significant impairments in daily functioning. Traditional exposure therapies, including Cognitive-Behavioral Therapy (CBT), have shown efficacy in treating SAD but are often limited by logistical challenges and patient reluctance. Virtual Reality Exposure Therapy (VRET) has emerged as a promising alternative, providing a controlled and immersive environment for exposure to anxiety-provoking scenarios. This paper explores the integration of Generative Adversarial Networks (GANs) into VRET to enhance the personalization, realism, and adaptability of therapeutic scenarios. GANs, which consist of a generator and a discriminator network, allow for the creation of highly realistic virtual environments tailored to the specific fears and triggers of individual patients. By offering real-time adjustments and personalized exposure scenarios, GAN-enhanced VRET addresses key limitations of traditional therapies, increasing patient engagement and treatment adherence. This review synthesizes findings from key studies on GAN-enhanced VRET, evaluating its efficacy in reducing anxiety symptoms and discussing the potential for broader application across various anxiety disorders. The results indicate that GAN-enhanced VRET offers a significant advancement in the treatment of SAD, providing a more accessible, engaging, and effective approach to exposure therapy. Future directions include expanding accessibility through cost reduction, integrating additional AI techniques, and exploring applications beyond SAD to improve mental health outcomes across diverse patient populations.

Image generation optimization is an important research direction in the field of deep learning, which aims to improve the performance of image generation models and the quality of generated images. In recent years, researchers have made significant progress in image generation optimization with the development of deep generative models such as generative adversarial networks (GANs) and variational autoencoders (VAEs). These models are able to generate high-quality, realistic images by learning the distribution of image data. In this study, a deep learning-based image generation optimization model was adopted, which combined the advantages of GAN and VAE. The model architecture consists of a generator and a discriminator, where the generator is responsible for generating the image and the discriminator is used to judge the authenticity of the image. In addition, the model also introduces attention mechanism and self-supervised learning strategy to further improve the quality and diversity of generated images. In the training process, a large-scale image dataset is used, and a variety of optimization algorithms are used to improve the stability and efficiency of the model. By evaluating various indicators of the generative model, including image quality, generation speed and model convergence, it was found that the introduced attention mechanism and self-supervised learning strategy significantly improved the performance of the model.

Neural or mental simulation of actions is a powerful tool for allowing cognitive agents to develop Prospection Capabilities that are crucial for learning and memorizing key aspects for challenging skills. In previous studies we developed an approach based on the animation of the redundant human Body Schema based on the Passive Motion Paradigm (PMP). In this paper we show that this approach can be easily extended to hyper-redundant serpentine robots as well as to hybrid configurations where the serpentine robot is functionally integrated with a traditional skeletal infrastructure. A simulation model is analyzed in detail showing that it incorporates spatio-temporal features discovered in the biomechanical studies of biological hydrostats like the elephant trunk or the octopus tentacles. It is proposed that such generative internal model can be the basis for a cognitive architecture appropriate for serpentine robots, independent of the underlying design and control technologies. Although robotic hydrostats have received a lot of attention in the last decades, the great majority of research activities have been focused on the actuation/sensorial/material technologies that can support the design of hyperredundant soft robots as well as the related control methodologies. The cognitive level of analysis has been limited to motion planning, without addressing synergy formation and metal time travel. This is where the contribution of the paper is focused on.

Although the statistical methods of downscaling climate data have progressed significantly, the development of high-resolution precipitation products continues to be a challenge. This is especially true when interest centres on downscaling value over several study sites. In this paper , we report a new downscaling method termed the multi-site Climate Generative Adversarial Network (MSCliGAN), which can simulate annual maximum precipitation to the regional scale during the 1950-2010 period in different cities in Canada by using different AOGCM's from the Coupled Model Inter-Comparison Project 6 (CMIP6) as input. Auxiliary information provided to the downscaling model included topography and land-cover. The downscaling framework uses a convolution encoder-decoder U-net network to create a generative network and a convolution encoder network to create a critic network. An adversarial training strategy is used to train the model. The critic/discriminator used Wasserstein distance as a loss measure and on the other hand the generator is optimized using a summation of content loss on Nash-Shutcliff Model Efficiency (NS), structural loss on structural similarity index (SSIM), and adversarial loss Wasserstein distance. Downscaling results show that downscaling AOGCMs by incorporating topography and land-use/land-cover can produce spatially coherent fields close to observation over multiple-sites. We believe the model has sufficient downscaling potential in data sparse regions where climate change information is often urgently needed.

This exploratory study investigates the use of Generative Artificial Intelligence (GenAI) tools in developing strategy practices for nanostores — i.e., small independent grocery retailers — to enhance their competitiveness while contributing to community sustainability. Nanostores, particularly in emerging countries, face intense competition and rapidly changing trends. By leveraging their proximity to consumers in neighbourhoods, these stores adopt various strategies, such as offering complementary services, integrating technology, and adapting business models. While existing literature highlights the broader economic, social, and environmental roles of nanostores, there is limited comprehensive research on how they may simultaneously contribute to their competitiveness and communities’ sustainability. This study employs a 'thing ethnography' approach, using GenAI tools —i.e., ChatGPT 3.5 and Microsoft Copilot— to explore potential strategy practices for these shops, and reflect on their possible use for strategy ideation and implementation. Key findings suggest that prioritising local sourcing, understanding local preferences, flexible inventory management, efficient retail operations, in-store sustainable practices, promoting healthy lifestyles, and diverse community engagement are crucial for success. However, the study also acknowledges the limitations of GenAI tools, including possible bias and the need for result validation. By aligning with these evolving requirements, nanostores can strive in neighbourhoods and decisively promote sustainable communities.

Art therapy fosters emotional healing and growth. This process can offer healthcare professionals (HCPs) novel insights into patients’ medication experiences. We developed a Metaverse Art Gal-lery of Image Chronicles (MAGIC), which depicted patients’ medication experiences symbolically as hero-villain portrayals. This gallery aimed to enhance healthcare students’ learning through re-latable insights into patients’ medication therapies. A character sheet was used to craft patients’ personifications of their medication experiences. ChatGPT, NightCafe, Canva, HeyGen and Cam-tasia were used to generate hero-villain portraits based on the character traits and mounted in MAGIC, which consisted of three virtual realms, each with a unique theme. Alpha-testing among sixteen Generation Z healthcare learners indicated that the content in MAGIC enabled them to understand the concepts of medication adherence (93.7%), art therapy (87.5%) and how patients related to their medications (81.3%). Perceived playfulness (rs=0.925, p

(1) Background: Artificial Intelligence (AI) models have shown potential in various educational contexts. However, their utility in explaining complex biological phenomena, such as Intrinsically Disordered Proteins (IDPs), requires further exploration. This study empirically evaluated the performance of various Large Language Models (LLMs) in the educational domain of IDPs. (2) Methods: Four LLMs, GPT-3.5, GPT-4, GPT-4 with Browsing, and Google Bard (PaLM 2), were assessed using a set of IDP-related questions. An expert evaluated their responses across five categories: accuracy, relevance, depth of understanding, clarity, and overall quality. Descriptive statistics, ANOVA, and Tukey's honesty significant difference tests were utilized for analysis. (3) Results: The GPT-4 model consistently outperformed the others across all evaluation categories. Although GPT-4 and GPT-3.5 were not statistically significantly different in performance (p>0.05), GPT-4 was preferred as the best response in 13 out of 15 instances. The AI models with browsing capabilities, GPT-4 with Browsing and Google Bard (PaLM 2) displayed lower performance metrics across the board with statistically significant differences (p<0.0001). (4) Conclusion: Our findings underscore the potential of AI models, particularly LLMs such as GPT-4, in enhancing scientific education, especially in complex domains such as IDPs. Continued innovation and collaboration among AI developers, educators, and researchers are essential to fully harness the potential of AI for enriching scientific education.

One of the most rapidly advancing areas of deep learning research aims at creating models that learn to disentangle the latent factors of variation from a data distribution. However, modeling joint probability mass functions is usually prohibitive, which motivates the use of conditional models assuming that some information is given as input. In the domain of numerical cognition, deep learning architectures have successfully demonstrated that approximate numerosity representations can emerge in multi-layer networks that build latent representations of a set of images with a varying number of items. However, existing models have focused on tasks requiring to conditionally estimate numerosity information from a given image. Here we focus on a set of much more challenging tasks, which require to conditionally generate synthetic images containing a given number of items. We show that attention-based architectures operating at the pixel level can learn to produce well-formed images approximately containing a specific number of items, even when the target numerosity was not present in the training distribution.

This paper examines the fundamental distinctions and complementary relationships between generative and descriptive methods in research and analysis. Through a systematic review of their applications across various fields, we explore how descriptive methods excel in capturing and characterizing existing phenomena, while generative methods enable the creation of new instances based on learned patterns. The analysis reveals that while these approaches serve different primary purposes, their integration often leads to more robust and comprehensive research outcomes. Our findings suggest that understanding the strengths and limitations of both methodologies is crucial for researchers and practitioners in choosing appropriate approaches for their specific contexts.

Conducting computational stress-strain analysis using finite element methods (FEM) is a common approach when dealing with the complex geometries of atherosclerosis, which is a leading cause of global mortality and complex cardiovascular disease. The considerable expense linked to FEM analysis encourages the substitution of FEM with a considerably faster data-driven machine learning (ML) approach. This study investigated the potential of end-to-end deep learning tools as a more effective substitute for FEM in predicting stress-strain fields within 2D cross sections of arterial wall. We first proposed a U-Net based fully convolutional neural network (CNN) to predict the von Mises stress and strain distribution based on the spatial arrangement of calcification within arterial wall cross-sections. Further, we developed a conditional generative adversarial network (cGAN) to enhance, particularly from the perceptual perspective, the prediction accuracy of stress and strain field maps for arterial walls with various calcification quantities and spatial configurations. On top of U-Net and cGAN, we also proposed their ensemble approaches to further improve the prediction accuracy of field maps. Our dataset, consisting of input and output images, was generated by implementing boundary conditions and extracting stress-strain field maps. The trained U-Net models can accurately predict von Mises stress and strain fields, with structural similarity index scores (SSIM) of 0.854 and 0.830 and mean squared errors of 0.017 and 0.018 for stress and strain, respectively, on a reserved test set. Meanwhile, the cGAN models in a combination of ensemble and transfer learning techniques demonstrate high accuracy in predicting von Mises stress and strain fields, as evidenced by SSIM scores of 0.890 for stress and 0.803 for strain. Additionally, mean squared errors of 0.008 for stress and 0.017 for strain further support the model's performance on a designated test set. Overall, this study developed a surrogate model for finite element analysis, which can accurately and efficiently predict stress-strain fields of arterial walls regardless of complex geometries and boundary conditions.

Diffusion Models (DMs) are probabilistic models that create realistic samples by simulating the diffusion process, gradually adding and removing noise from data. These models have gained popularity in domains such as image processing, speech synthesis, and natural language processing due to their ability to produce high-quality samples. As DMs are being adopted in various domains, existing literature reviews that often focus on specific areas like computer vision or medical imaging may not serve a broader audience across multiple fields. Therefore, this review presents a comprehensive overview of DMs, covering their theoretical foundations and algorithmic innovations. We highlight their applications in diverse areas such as media quality, authenticity, synthesis, image transformation, healthcare, and more. By consolidating current knowledge and identifying emerging trends, this review aims to facilitate a deeper understanding and broader adoption of DMs and provide guidelines for future researchers and practitioners across diverse disciplines.

This research investigates the potential application of generative language models, especially ChatGPT, in aviation safety analysis as a means to enhance the efficiency of safety analyses and accelerate the time it takes to process incident reports. In particular, ChatGPT was leveraged to generate incident synopses from narratives, which were subsequently compared with ground truth synopses from the Aviation Safety Reporting System (ASRS) dataset. The comparison was facilitated by using embeddings from Large Language Models (LLMs), with aeroBERT demonstrating the highest similarity due to its aerospace-specific fine-tuning. A positive correlation was observed between synopsis length and their cosine similarity. In a subsequent phase, human factor issues involved in incidents as identified by ChatGPT were compared to human factor issues identified by safety analysts. A concurrence rate of 61% was found, with ChatGPT demonstrating a cautious approach towards attributing human factor issues. Finally, the model was used to attribute incidents to relevant parties. As no dedicated ground truth column existed for this task, a manual evaluation was conducted. ChatGPT attributed the majority of incidents to the Flight Crew, ATC, Ground Personnel, and Maintenance. This study opens new avenues for leveraging AI in aviation safety analysis.

Background: Open AI developed ChatGPT, a language model based on the GPT architecture, designed for text-based communication. Trained on diverse internet texts, ChatGPT generates contextually appropriate responses using machine learning. It understands input through analysis and context interpretation, generating coherent and contextually relevant responses. Interaction is possible through messaging platforms. Materials and Methods: In November 2023, we utilized ChatGPT 3.5, the default version at that time, to answer questions from the Italian National Residency Program Admission Tests (SSMs) of 2021, 2022, and 2023. The questions cover clinical, diagnostic, analytical, therapeutic, and epidemiological scenarios, sometimes accompanied by images. The study compared ChatGPT's answers to the official corrections on the Italian Ministry of University and Research (MUR) website. The scoring method used for evaluation was 1 point for correct answers, 0 points for unanswered questions, and -0.25 points for incorrect answers, reflecting the SSM test scoring system. Results: In summary, ChatGPT was tested with a total of 420 questions, 140 for each test. It achieved an overall accuracy of 80.48%, providing correct answers for 338 questions and incorrect answers for 82 questions. When faced with questions containing both text and images, it answered 55% correctly and 45% incorrectly. The model's performance varied over time, with an 82.14% accuracy rate in 2021 and 2022 (115 correct out of 140) and a 77.14% accuracy rate in 2023 (108 correct out of 140). Applying thisscoring method to the SSM test, ChatGPT would have scored 105 points in 2021 and 2022, and 100 points in 2023. Conclusions: ChatGPT has exhibited above-average performance in the last three SSM tests, highlighting its robust capability to interpret clinical scenarios and offer precise diagnostic and therapeutic guidance. Despite this, some limitations persist, notably the software's inability to interpret non-textual information.

Background: This study addresses the burgeoning impact of Large Language Models (LLMs) such as ChatGPT on the integrity of online examinations and seeks to demonstrate latent and advanced LLM reasoning capabilities. The research devises an iterative self-reflective strategy for invoking critical thinking and higher-order reasoning in LLMs based on multimodal exam questions using visuals and text. Methods: The proposed strategy was demonstrated and evaluated on real exam questions by subject experts and the performance of ChatGPT (GPT-4) with vision was estimated on an additional dataset of 600 text descriptions of multimodal exam questions, focusing on its response to complex, visual, and textual prompts. Results: The results indicate that latent multi-hop reasoning capabilities can be invoked using the proposed self-reflective strategy, which can effectively steer ChatGPT towards correct answers by integrating critical thinking from each modality into the final response. Meanwhile, ChatGPT demonstrated considerable proficiency in being able to answer multimodal exam questions across 12 subjects. Conclusions: This work challenges findings from recent studies which claim that existing LLMs are unable to reason on complex multimodal tasks, and therefore this study underscores the need for enhanced online exam security measures like proctoring software, that can mitigate the potential misuse of AI technologies in educational settings.

Generative AI chatbots have emerged as a significant scientific contribution. They can produce text, images, audio, and video, and their applications are vast and varied in every field. However, it is identified that these chatbots can be used by script kiddies and malicious actors to generate phishing emails, security exploits, and executable payloads that can present notable risks and challenges for cybersecurity. This research provides an extensive review of the current status of the three most popular generative AI chatbots which are ChatGPT, Google Gemini, and Microsoft Bing. The article further reviews the different ways in which an attacker can intentionally use a chatbot for malicious activities and intensify cyber attacks. Moreover, the article provides insights on how an attacker uses certain keyword queries to manipulate chatbot behavior and tricks it into generating content that I won’t otherwise. We also explored the role of AI chatbots in enhancing cyber resilience against sophisticated attacks and their applications in detecting and mitigating security incidents. Finally, we proposed strategies that can empower modern chatbots to defend against such threats so that an attacker will not be able to bypass their content filtering mechanisms.

This research delves into the intricate challenges confronting the agricultural sector, with a specialized focus on mitigating infections in tomato crops, particularly powdery mildew induced by the Leveillula Taurica pathogen. Tomatoes, renowned for their nutritional richness, are vital to global food security. However, conventional methodologies for disease detection exhibit both laborious processes and limited accuracy. In response to these challenges, this study advocates for an innovative fusion of hyperspectral imaging and deep learning methodologies to detect crop disease and its severity. The systematic workflow commenced with the curation of a dataset, involving the acquisition of live images through OpenCV, followed by conversion to RGB format and subsequent feature extraction utilizing a pre-trained visual geometry group (VGG-16) model for enhanced analysis. Sequentially, RGB images were transformed into simulated hyperspectral images (SHSI) leveraging a Neural Network generator model, offering a distinctive viewpoint on spectral information. This novel approach transcends conventional constraints by delivering a three-dimensional perspective, seamlessly integrating spatial and spectral dimensions for holistic data acquisition. The SHSI is further transmuted into a 3D visualization cube comprehensive grasp of spatial and spectral aspects encompassing spectral, spatial, and Haralick features. The research concludes with severity detection, categorized as low, moderate, or high, employing a Gaussian Mixture Model (GMM) and K-means for visualization.