Integrated sensing and communications(ISAC)is a key enabler for next-generation wireless systems,aiming to support both high-throughput communication and high-accuracy environmental sensing using shared spectrum and h...Integrated sensing and communications(ISAC)is a key enabler for next-generation wireless systems,aiming to support both high-throughput communication and high-accuracy environmental sensing using shared spectrum and hardware.Theoretical performance metrics,such as mutual information(MI),minimum mean squared error(MMSE),and Bayesian Cram´er-Rao bound(BCRB),play a key role in evaluating ISAC system performance limits.However,in practice,hardware impairments,multipath propagation,interference,and scene constraints often result in nonlinear,multimodal,and non-Gaussian distributions,making it challenging to derive these metrics analytically.Recently,there has been a growing interest in applying score-based generative models to characterize these metrics from data,although not discussed for ISAC.This paper provides a tutorial-style summary of recent advances in score-based performance evaluation,with a focus on ISAC systems.We refer to the summarized framework as scoring ISAC,which not only reflects the core methodology based on score functions but also emphasizes the goal of scoring(i.e.,evaluating)ISAC systems under realistic conditions.We present the connections between classical performance metrics and the score functions and provide the practical training techniques for learning score functions to estimate performance metrics.Proof-of-concept experiments on target detection and localization validate the accuracy of score-based performance estimators against groundtruth analytical expressions,illustrating their ability to replicate and extend traditional analyses in more complex,realistic settings.This framework demonstrates the great potential of score-based generative models in ISAC performance analysis,algorithm design,and system optimization.展开更多
Over the past century,advancements in chemistry have significantly propelled human innovation,enhancing both industrial and consumer products.However,this rapid progression has resulted in chemical pollution increasin...Over the past century,advancements in chemistry have significantly propelled human innovation,enhancing both industrial and consumer products.However,this rapid progression has resulted in chemical pollution increasingly surpassing planetary boundaries,as production and release rates have outpaced our monitoring capabilities.To catalyze more impactful efforts,this study transitions from traditional chemical assessment to inverse chemical design,introducing a generative graph latent diffusion model aimed at discovering safer alternatives.In a case study on the design of green solvents for cyclohexane/benzene extraction distillation,we constructed a design database encompassing functional,environmental hazards,and process constraints.Virtual screening of previous design dataset revealed distinct trade-off trends between these design requirements.Based on the screening outcomes,an unconstrained generative model was developed,which covered a broader chemical space and demonstrated superior capabilities for structural interpolation and extrapolation.To further optimize molecular generation towards desired properties,a multi-objective latent diffusion method was applied,yielding 19 candidate molecules.Of these,7 were identified in PubChem as the most viable green solvent candidates,while the remaining 12 as potential novel candidates.Overall,this study effectively designed green solvent candidates for safer and more sustainable industrial production,setting a promising precedent for the development of environmentally friendly alternatives in other areas of chemical research.展开更多
Short-term building energy predictions serve as one of the fundamental tasks in building operation management.While large numbers of studies have explored the value of various supervised machine learning techniques in...Short-term building energy predictions serve as one of the fundamental tasks in building operation management.While large numbers of studies have explored the value of various supervised machine learning techniques in energy predictions,few studies have addressed the potential data shortage problem in developing data-driven models.One promising solution is data augmentation,which aims to enrich existing building data resources for reliable predictive modeling.This study proposes a deep generative modeling-based data augmentation strategy for improving short-term building energy predictions.Two types of conditional variational autoencoders have been designed for synthetic energy data generation using fully connected and one-dimensional convolutional layers respectively.Data experiments have been designed to evaluate the value of data augmentation using actual measurements from 52 buildings.The results indicate that conditional variational autoencoders are capable of generating high-quality synthetic data samples,which in turns helps to enhance the accuracy in short-term building energy predictions.The average performance enhancement ratios in terms of CV-RMSE range between 12%and 18%.Practical guidelines have been obtained to ensure the validity and quality of synthetic building energy data.The research outcomes are valuable for enhancing the robustness and reliability of data-driven models for smart building operation management.展开更多
Robust stereo disparity estimation plays a critical role in minimally invasive surgery,where dynamic soft tissues,specular reflections,and data scarcity pose major challenges to traditional end-to-end deep learning an...Robust stereo disparity estimation plays a critical role in minimally invasive surgery,where dynamic soft tissues,specular reflections,and data scarcity pose major challenges to traditional end-to-end deep learning and deformable model-based methods.In this paper,we propose a novel disparity estimation framework that leverages a pretrained StyleGAN generator to represent the disparity manifold of Minimally Invasive Surgery(MIS)scenes and reformulates the stereo matching task as a latent-space optimization problem.Specifically,given a stereo pair,we search for the optimal latent vector in the intermediate latent space of StyleGAN,such that the photometric reconstruction loss between the stereo images is minimized while regularizing the latent code to remain within the generator’s high-confidence region.Unlike existing encoder-based embedding methods,our approach directly exploits the geometry of the learned latent space and enforces both photometric consistency and manifold prior during inference,without the need for additional training or supervision.Extensive experiments on stereo-endoscopic videos demonstrate that our method achieves high-fidelity and robust disparity estimation across varying lighting,occlusion,and tissue dynamics,outperforming Thin Plate Spline(TPS)-based and linear representation baselines.This work bridges generative modeling and 3D perception by enabling efficient,training-free disparity recovery from pre-trained generative models with reduced inference latency.展开更多
Recently,diffusion models have emerged as a promising paradigm for molecular design and optimization.However,most diffusion-based molecular generative models focus on modeling 2D graphs or 3D geom-etries,with limited ...Recently,diffusion models have emerged as a promising paradigm for molecular design and optimization.However,most diffusion-based molecular generative models focus on modeling 2D graphs or 3D geom-etries,with limited research on molecular sequence diffusion models.The International Union of Pure and Applied Chemistry(IUPAC)names are more akin to chemical natural language than the simplified molecular input line entry system(SMILES)for organic compounds.In this work,we apply an IUPAC-guided conditional diffusion model to facilitate molecular editing from chemical natural language to chemical language(SMILES)and explore whether the pre-trained generative performance of diffusion models can be transferred to chemical natural language.We propose DiffIUPAC,a controllable molecular editing diffusion model that converts IUPAC names to SMILES strings.Evaluation results demonstrate that our model out-performs existing methods and successfully captures the semantic rules of both chemical languages.Chemical space and scaffold analysis show that the model can generate similar compounds with diverse scaffolds within the specified constraints.Additionally,to illustrate the model’s applicability in drug design,we conducted case studies in functional group editing,analogue design and linker design.展开更多
The exponential growth of over-the-top(OTT)entertainment has fueled a surge in content consumption across diverse formats,especially in regional Indian languages.With the Indian film industry producing over 1500 films...The exponential growth of over-the-top(OTT)entertainment has fueled a surge in content consumption across diverse formats,especially in regional Indian languages.With the Indian film industry producing over 1500 films annually in more than 20 languages,personalized recommendations are essential to highlight relevant content.To overcome the limitations of traditional recommender systems-such as static latent vectors,poor handling of cold-start scenarios,and the absence of uncertainty modeling-we propose a deep Collaborative Neural Generative Embedding(C-NGE)model.C-NGE dynamically learns user and item representations by integrating rating information and metadata features in a unified neural framework.It uses metadata as sampled noise and applies the reparameterization trick to capture latent patterns better and support predictions for new users or items without retraining.We evaluate CNGE on the Indian Regional Movies(IRM)dataset,along with MovieLens 100 K and 1 M.Results show that our model consistently outperforms several existing methods,and its extensibility allows for incorporating additional signals like user reviews and multimodal data to enhance recommendation quality.展开更多
The growing demand for wireless connectivity has made massive multiple-input multiple-output(MIMO)a cornerstone of modern communication systems.To optimize network performance and resource allocation,an efficient and ...The growing demand for wireless connectivity has made massive multiple-input multiple-output(MIMO)a cornerstone of modern communication systems.To optimize network performance and resource allocation,an efficient and robust approach is joint device activity detection and channel estimation.In this paper,we present an approach utilizing score-based generative models to address the underdetermined nature of channel estimation,which is data-driven and well-suited for the complex and dynamic environment of massive MIMO systems.Our experimental results,based on a comprehensive dataset generated through Monte-Carlo sampling,demonstrate the high precision of our channel estimation approach,with errors reduced to as low as-45 d B,and exceptional accuracy in detecting active devices.展开更多
The increasing frequency of extreme weather events raises the likelihood of forest wildfires.Therefore,establishing an effective fire prediction model is vital for protecting human life and property,and the environmen...The increasing frequency of extreme weather events raises the likelihood of forest wildfires.Therefore,establishing an effective fire prediction model is vital for protecting human life and property,and the environment.This study aims to build a prediction model to understand the spatial characteristics and piecewise effects of forest fire drivers.Using monthly grid data from 2006 to 2020,a modeling study analyzed fire occurrences during the September to April fire season in Fujian Province,China.We compared the fitting performance of the logistic regression model(LRM),the generalized additive logistic model(GALM),and the spatial generalized additive logistic model(SGALM).The results indicate that SGALMs had the best fitting results and the highest prediction accuracy.Meteorological factors significantly impacted forest fires in Fujian Province.Areas with high fire incidence were mainly concentrated in the northwest and southeast.SGALMs improved the fitting effect of fire prediction models by considering spatial effects and the flexible fitting ability of nonlinear interpretation.This model provides piecewise interpretations of forest wildfire occurrences,which can be valuable for relevant departments and will assist forest managers in refining prevention measures based on temporal and spatial differences.展开更多
Generative modeling(GM)has been increasingly used for the inverse design and optimization of materials,yet its application to molten salt mixtures remains unexplored despite how a successful approach to the inverse de...Generative modeling(GM)has been increasingly used for the inverse design and optimization of materials,yet its application to molten salt mixtures remains unexplored despite how a successful approach to the inverse design of molten salts would contribute to efficiently exploiting their customizability and unlocking their advantages in applications,such as energy production and energy storage.This work presents a workflow for the inverse design of molten salts with targeted density values,addressing the challenge of representing these complex mixtures inGM.A dataset of critically evaluated molten salt densities is used to train a variational autoencoder coupled with a predictive deep neural network,which then can be used to generate new molten salt compositions with desired density values.The effectiveness of the approach is demonstrated by designing mixtures with distinct densities and validating the predicted values using ab initio molecular dynamics simulations.展开更多
Nanomaterials’properties,influenced by size,shape,and surface characteristics,are crucial for their technological,biological,and environmental applications.Accurate quantification of these materials is essential for ...Nanomaterials’properties,influenced by size,shape,and surface characteristics,are crucial for their technological,biological,and environmental applications.Accurate quantification of these materials is essential for advancing research.Deep learning segmentation networks offer precise,automated analysis,but their effectiveness depends on representative annotated datasets,which are difficult to obtain due to the high cost and manual effort required for imaging and annotation.To address this,we present DiffRenderGAN,a generative model that produces annotated synthetic data by integrating a differentiable renderer into a Generative Adversarial Network(GAN)framework.DiffRenderGAN optimizes rendering parameters to produce realistic,annotated images from non-annotated real microscopy images,reducing manual effort and improving segmentation performance compared to existing methods.Tested on ion and electron microscopy datasets,including titanium dioxide(TiO_(2)),silicon dioxide(SiO_(2)),and silver nanowires(AgNW),DiffRenderGAN bridges the gap between synthetic and real data,advancing the quantification and understanding of complex nanomaterial systems.展开更多
Recently, generative adversarial networks(GANs)have become a research focus of artificial intelligence. Inspired by two-player zero-sum game, GANs comprise a generator and a discriminator, both trained under the adver...Recently, generative adversarial networks(GANs)have become a research focus of artificial intelligence. Inspired by two-player zero-sum game, GANs comprise a generator and a discriminator, both trained under the adversarial learning idea.The goal of GANs is to estimate the potential distribution of real data samples and generate new samples from that distribution.Since their initiation, GANs have been widely studied due to their enormous prospect for applications, including image and vision computing, speech and language processing, etc. In this review paper, we summarize the state of the art of GANs and look into the future. Firstly, we survey GANs' proposal background,theoretic and implementation models, and application fields.Then, we discuss GANs' advantages and disadvantages, and their development trends. In particular, we investigate the relation between GANs and parallel intelligence,with the conclusion that GANs have a great potential in parallel systems research in terms of virtual-real interaction and integration. Clearly, GANs can provide substantial algorithmic support for parallel intelligence.展开更多
Inverse design has long been an efficient and powerful design tool in the aircraft industry.In this paper,a novel inverse design method for supercritical airfoils is proposed based on generative models in deep learnin...Inverse design has long been an efficient and powerful design tool in the aircraft industry.In this paper,a novel inverse design method for supercritical airfoils is proposed based on generative models in deep learning.A Conditional Variational Auto Encoder(CVAE)and an integrated generative network CVAE-GAN that combines the CVAE with the Wasserstein Generative Adversarial Networks(WGAN),are conducted as generative models.They are used to generate target wall Mach distributions for the inverse design that matches specified features,such as locations of suction peak,shock and aft loading.Qualitative and quantitative results show that both adopted generative models can generate diverse and realistic wall Mach number distributions satisfying the given features.The CVAE-GAN model outperforms the CVAE model and achieves better reconstruction accuracies for all the samples in the dataset.Furthermore,a deep neural network for nonlinear mapping is adopted to obtain the airfoil shape corresponding to the target wall Mach number distribution.The performances of the designed deep neural network are fully demonstrated and a smoothness measurement is proposed to quantify small oscillations in the airfoil surface,proving the authenticity and accuracy of the generated airfoil shapes.展开更多
For reservoirs with complex non-Gaussian geological characteristics,such as carbonate reservoirs or reservoirs with sedimentary facies distribution,it is difficult to implement history matching directly,especially for...For reservoirs with complex non-Gaussian geological characteristics,such as carbonate reservoirs or reservoirs with sedimentary facies distribution,it is difficult to implement history matching directly,especially for the ensemble-based data assimilation methods.In this paper,we propose a multi-source information fused generative adversarial network(MSIGAN)model,which is used for parameterization of the complex geologies.In MSIGAN,various information such as facies distribution,microseismic,and inter-well connectivity,can be integrated to learn the geological features.And two major generative models in deep learning,variational autoencoder(VAE)and generative adversarial network(GAN)are combined in our model.Then the proposed MSIGAN model is integrated into the ensemble smoother with multiple data assimilation(ESMDA)method to conduct history matching.We tested the proposed method on two reservoir models with fluvial facies.The experimental results show that the proposed MSIGAN model can effectively learn the complex geological features,which can promote the accuracy of history matching.展开更多
Habitat suitability index(HSI)models have been widely used to analyze the relationship between species abundance and environmental factors,and ultimately inform management of marine species.The response of species abu...Habitat suitability index(HSI)models have been widely used to analyze the relationship between species abundance and environmental factors,and ultimately inform management of marine species.The response of species abundance to each environmental variable is different and habitat requirements may change over life history stages and seasons.Therefore,it is necessary to determine the optimal combination of environmental variables in HSI modelling.In this study,generalized additive models(GAMs)were used to determine which environmental variables to be included in the HSI models.Significant variables were retained and weighted in the HSI model according to their relative contribution(%)to the total deviation explained by the boosted regression tree(BRT).The HSI models were applied to evaluate the habitat suitability of mantis shrimp Oratosquilla oratoria in the Haizhou Bay and adjacent areas in 2011 and 2013–2017.Ontogenetic and seasonal variations in HSI models of mantis shrimp were also examined.Among the four models(non-optimized model,BRT informed HSI model,GAM informed HSI model,and both BRT and GAM informed HSI model),both BRT and GAM informed HSI model showed the best performance.Four environmental variables(bottom temperature,depth,distance offshore and sediment type)were selected in the HSI models for four groups(spring-juvenile,spring-adult,falljuvenile and fall-adult)of mantis shrimp.The distribution of habitat suitability showed similar patterns between juveniles and adults,but obvious seasonal variations were observed.This study suggests that the process of optimizing environmental variables in HSI models improves the performance of HSI models,and this optimization strategy could be extended to other marine organisms to enhance the understanding of the habitat suitability of target species.展开更多
In this paper,we propose a novel coverless image steganographic scheme based on a generative model.In our scheme,the secret image is first fed to the generative model database,to generate a meaning-normal and independ...In this paper,we propose a novel coverless image steganographic scheme based on a generative model.In our scheme,the secret image is first fed to the generative model database,to generate a meaning-normal and independent image different from the secret image.The generated image is then transmitted to the receiver and fed to the generative model database to generate another image visually the same as the secret image.Thus,we only need to transmit the meaning-normal image which is not related to the secret image,and we can achieve the same effect as the transmission of the secret image.This is the first time to propose the coverless image information steganographic scheme based on generative model,compared with the traditional image steganography.The transmitted image is not embedded with any information of the secret image in this method,therefore,can effectively resist steganalysis tools.Experimental results show that our scheme has high capacity,security and reliability.展开更多
This research develops a new mathematical modeling method by combining industrial big data and process mechanism analysis under the framework of generalized additive models(GAM)to generate a practical model with gener...This research develops a new mathematical modeling method by combining industrial big data and process mechanism analysis under the framework of generalized additive models(GAM)to generate a practical model with generalization and precision.Specifically,the proposed modeling method includes the following steps.Firstly,the influence factors are screened using mechanism knowledge and data-mining methods.Secondly,the unary GAM without interactions including cleaning the data,building the sub-models,and verifying the sub-models.Subsequently,the interactions between the various factors are explored,and the binary GAM with interactions is constructed.The relationships among the sub-models are analyzed,and the integrated model is built.Finally,based on the proposed modeling method,two prediction models of mechanical property and deformation resistance for hot-rolled strips are established.Industrial actual data verification demonstrates that the new models have good prediction precision,and the mean absolute percentage errors of tensile strength,yield strength and deformation resistance are 2.54%,3.34%and 6.53%,respectively.And experimental results suggest that the proposed method offers a new approach to industrial process modeling.展开更多
Predictive models for assessing the risk of developing lung cancers can help identify high-risk individuals with the aim of recommending further screening and early intervention.To facilitate pre-hospital self-assessm...Predictive models for assessing the risk of developing lung cancers can help identify high-risk individuals with the aim of recommending further screening and early intervention.To facilitate pre-hospital self-assessments,some studies have exploited predictive models trained on non-clinical data(e.g.,smoking status and family history).The performance of these models is limited due to not considering clinical data(e.g.,blood test and medical imaging results).Deep learning has shown the potential in processing complex data that combine both clinical and non-clinical information.However,predicting lung cancers remains difficult due to the severe lack of positive samples among follow-ups.To tackle this problem,this paper presents a generative-discriminative framework for improving the ability of deep learning models to generalize.According to the proposed framework,two nonlinear generative models,one based on the generative adversarial network and another on the variational autoencoder,are used to synthesize auxiliary positive samples for the training set.Then,several discriminative models,including a deep neural network(DNN),are used to assess the lung cancer risk based on a comprehensive list of risk factors.The framework was evaluated on over 55000 subjects questioned between January 2014 and December 2017,with 699 subjects being clinically diagnosed with lung cancer between January 2014 and August 2019.According to the results,the best performing predictive model built using the proposed framework was based on DNN.It achieved an average sensitivity of 76.54%and an area under the curve of 69.24%in distinguishing between the cases of lung cancer and normal cases on test sets.展开更多
Natural products（NPs） have long been recognized as a valuable resource for drug discovery, and bringing NP-related features to virtual libraries is believed to be an effective way to increase the coverage of druggab...Natural products（NPs） have long been recognized as a valuable resource for drug discovery, and bringing NP-related features to virtual libraries is believed to be an effective way to increase the coverage of druggable chemical space. Here, deep learning-based molecule generative model, which is a recent technique in de novo molecule design, was applied to generate virtual libraries with NP-like properties. Results demonstrated that the model was effective in generating molecules that highly resemble NPs. Moreover, the model was also found to be capable of generating NP-like molecules that were also easy to synthesize, significantly increasing the practical value of the compound library.展开更多
Stochastic models are derived to estimate the level of coliform count in terms of MPN index, one of the most important water quality characteristic in ground water based on a set of water source location and soil char...Stochastic models are derived to estimate the level of coliform count in terms of MPN index, one of the most important water quality characteristic in ground water based on a set of water source location and soil characteristics. The study is based on about twenty location and soil characteristics, majority of them are observed through laboratory analysis of soil and water samples collected from nearly thee hundred locations of drinking water sources, wells and bore wells selected at random from the district of Kasaragod. The water contamination in wells are found to be relatively more as compared to bore wells. The study reveals that only 7 % of the wells and 40 o~ of the bore wells of the district are within the permissible limit of WHO standard of drinking water quality. The level of contamination is very high in the hospital premises and is very low in the forest area. Two separate multiple ordinal logistic regression models are developed to predict the level of coliform count, one for well and the other for bore well. The significant feature of this study is that in addition to scientifically proving the dependence of the water quality on the distances from waste disposal area and septic tanks etc., it highlights the dependence of two other very significant soil characteristics, the soil organic carbon and soil porosity. The models enable to predict the quality of water in a location based on the set of soil and location characteristics. One of the important uses of the model is in fixing safe locations for waste dump area, septic tank, digging well etc. in town planning, designing residential layouts, industrial layouts, hospital/hostel construction etc. This is the first ever study to describe the ground water quality in terms of the location and soil characteristics.展开更多
A model of deformation resistance during hot strip rolling was established based on generalized additive model.Firstly,a data modeling method based on generalized additive model was given.It included the selection of ...A model of deformation resistance during hot strip rolling was established based on generalized additive model.Firstly,a data modeling method based on generalized additive model was given.It included the selection of dependent variable and independent variables of the model,the link function of dependent variable and smoothing functional form of each independent variable,estimating process of the link function and smooth functions,and the last model modification.Then,the practical modeling test was carried out based on a large amount of hot rolling process data.An integrated variable was proposed to reflect the effects of different chemical compositions such as carbon,silicon,manganese,nickel,chromium,niobium,etc.The integrated chemical composition,strain,strain rate and rolling temperature were selected as independent variables and the cubic spline as the smooth function for them.The modeling process of deformation resistance was realized by SAS software,and the influence curves of the independent variables on deformation resistance were obtained by local scoring algorithm.Some interesting phenomena were found,for example,there is a critical value of strain rate,and the deformation resistance increases before this value and then decreases.The results confirm that the new model has higher prediction accuracy than traditional ones and is suitable for carbon steel,microalloyed steel,alloyed steel and other steel grades.展开更多
基金supported in part by the General Research Fund from the Research Grants Council of Hong Kong,China,under Project 14202421,Project 14214122,Project 14202723,and Project 14207624in part by Area of Excellence Scheme grant from the Research Grants Council of Hong Kong,China,under Project number AoE/E-601/22-R+2 种基金in part by NSFC/RGC Collaborative Research Scheme from the Research Grants Council of Hong Kong,China,under Project CRS HKUST603/22 and Project CRS HKU702/24in part by the National Natural Science Foundation of China under Grant 62571087in part by Sichuan Science and Technology Program under Grant 2024ZYD0036.
文摘Integrated sensing and communications(ISAC)is a key enabler for next-generation wireless systems,aiming to support both high-throughput communication and high-accuracy environmental sensing using shared spectrum and hardware.Theoretical performance metrics,such as mutual information(MI),minimum mean squared error(MMSE),and Bayesian Cram´er-Rao bound(BCRB),play a key role in evaluating ISAC system performance limits.However,in practice,hardware impairments,multipath propagation,interference,and scene constraints often result in nonlinear,multimodal,and non-Gaussian distributions,making it challenging to derive these metrics analytically.Recently,there has been a growing interest in applying score-based generative models to characterize these metrics from data,although not discussed for ISAC.This paper provides a tutorial-style summary of recent advances in score-based performance evaluation,with a focus on ISAC systems.We refer to the summarized framework as scoring ISAC,which not only reflects the core methodology based on score functions but also emphasizes the goal of scoring(i.e.,evaluating)ISAC systems under realistic conditions.We present the connections between classical performance metrics and the score functions and provide the practical training techniques for learning score functions to estimate performance metrics.Proof-of-concept experiments on target detection and localization validate the accuracy of score-based performance estimators against groundtruth analytical expressions,illustrating their ability to replicate and extend traditional analyses in more complex,realistic settings.This framework demonstrates the great potential of score-based generative models in ISAC performance analysis,algorithm design,and system optimization.
基金supported by Shanghai Science and Technology Commission Project(No.21DZ1201502)Shanghai Municipal Bureau of Ecology and Environment(Shanghai Environ-mental Science[2023]No.40)+1 种基金the Interdisciplinary Joint Research Project of Tongji University(No.2022-4-YB-12)Shanghai Science and Technology Commission Project(No.22DZ2200200).
文摘Over the past century,advancements in chemistry have significantly propelled human innovation,enhancing both industrial and consumer products.However,this rapid progression has resulted in chemical pollution increasingly surpassing planetary boundaries,as production and release rates have outpaced our monitoring capabilities.To catalyze more impactful efforts,this study transitions from traditional chemical assessment to inverse chemical design,introducing a generative graph latent diffusion model aimed at discovering safer alternatives.In a case study on the design of green solvents for cyclohexane/benzene extraction distillation,we constructed a design database encompassing functional,environmental hazards,and process constraints.Virtual screening of previous design dataset revealed distinct trade-off trends between these design requirements.Based on the screening outcomes,an unconstrained generative model was developed,which covered a broader chemical space and demonstrated superior capabilities for structural interpolation and extrapolation.To further optimize molecular generation towards desired properties,a multi-objective latent diffusion method was applied,yielding 19 candidate molecules.Of these,7 were identified in PubChem as the most viable green solvent candidates,while the remaining 12 as potential novel candidates.Overall,this study effectively designed green solvent candidates for safer and more sustainable industrial production,setting a promising precedent for the development of environmentally friendly alternatives in other areas of chemical research.
基金support of this research by the National Natural Science Foundation of China(No.51908365,No.71772125)the Philosophical and Social Science Program of Guangdong Province,China(GD18YGL07).
文摘Short-term building energy predictions serve as one of the fundamental tasks in building operation management.While large numbers of studies have explored the value of various supervised machine learning techniques in energy predictions,few studies have addressed the potential data shortage problem in developing data-driven models.One promising solution is data augmentation,which aims to enrich existing building data resources for reliable predictive modeling.This study proposes a deep generative modeling-based data augmentation strategy for improving short-term building energy predictions.Two types of conditional variational autoencoders have been designed for synthetic energy data generation using fully connected and one-dimensional convolutional layers respectively.Data experiments have been designed to evaluate the value of data augmentation using actual measurements from 52 buildings.The results indicate that conditional variational autoencoders are capable of generating high-quality synthetic data samples,which in turns helps to enhance the accuracy in short-term building energy predictions.The average performance enhancement ratios in terms of CV-RMSE range between 12%and 18%.Practical guidelines have been obtained to ensure the validity and quality of synthetic building energy data.The research outcomes are valuable for enhancing the robustness and reliability of data-driven models for smart building operation management.
基金Support by Sichuan Science and Technology Program[2023YFSY0026,2023YFH0004]Guangzhou Huashang University[2024HSZD01].
文摘Robust stereo disparity estimation plays a critical role in minimally invasive surgery,where dynamic soft tissues,specular reflections,and data scarcity pose major challenges to traditional end-to-end deep learning and deformable model-based methods.In this paper,we propose a novel disparity estimation framework that leverages a pretrained StyleGAN generator to represent the disparity manifold of Minimally Invasive Surgery(MIS)scenes and reformulates the stereo matching task as a latent-space optimization problem.Specifically,given a stereo pair,we search for the optimal latent vector in the intermediate latent space of StyleGAN,such that the photometric reconstruction loss between the stereo images is minimized while regularizing the latent code to remain within the generator’s high-confidence region.Unlike existing encoder-based embedding methods,our approach directly exploits the geometry of the learned latent space and enforces both photometric consistency and manifold prior during inference,without the need for additional training or supervision.Extensive experiments on stereo-endoscopic videos demonstrate that our method achieves high-fidelity and robust disparity estimation across varying lighting,occlusion,and tissue dynamics,outperforming Thin Plate Spline(TPS)-based and linear representation baselines.This work bridges generative modeling and 3D perception by enabling efficient,training-free disparity recovery from pre-trained generative models with reduced inference latency.
基金supported by the Yonsei University graduate school Department of Integrative Biotechnology.
文摘Recently,diffusion models have emerged as a promising paradigm for molecular design and optimization.However,most diffusion-based molecular generative models focus on modeling 2D graphs or 3D geom-etries,with limited research on molecular sequence diffusion models.The International Union of Pure and Applied Chemistry(IUPAC)names are more akin to chemical natural language than the simplified molecular input line entry system(SMILES)for organic compounds.In this work,we apply an IUPAC-guided conditional diffusion model to facilitate molecular editing from chemical natural language to chemical language(SMILES)and explore whether the pre-trained generative performance of diffusion models can be transferred to chemical natural language.We propose DiffIUPAC,a controllable molecular editing diffusion model that converts IUPAC names to SMILES strings.Evaluation results demonstrate that our model out-performs existing methods and successfully captures the semantic rules of both chemical languages.Chemical space and scaffold analysis show that the model can generate similar compounds with diverse scaffolds within the specified constraints.Additionally,to illustrate the model’s applicability in drug design,we conducted case studies in functional group editing,analogue design and linker design.
文摘The exponential growth of over-the-top(OTT)entertainment has fueled a surge in content consumption across diverse formats,especially in regional Indian languages.With the Indian film industry producing over 1500 films annually in more than 20 languages,personalized recommendations are essential to highlight relevant content.To overcome the limitations of traditional recommender systems-such as static latent vectors,poor handling of cold-start scenarios,and the absence of uncertainty modeling-we propose a deep Collaborative Neural Generative Embedding(C-NGE)model.C-NGE dynamically learns user and item representations by integrating rating information and metadata features in a unified neural framework.It uses metadata as sampled noise and applies the reparameterization trick to capture latent patterns better and support predictions for new users or items without retraining.We evaluate CNGE on the Indian Regional Movies(IRM)dataset,along with MovieLens 100 K and 1 M.Results show that our model consistently outperforms several existing methods,and its extensibility allows for incorporating additional signals like user reviews and multimodal data to enhance recommendation quality.
文摘The growing demand for wireless connectivity has made massive multiple-input multiple-output(MIMO)a cornerstone of modern communication systems.To optimize network performance and resource allocation,an efficient and robust approach is joint device activity detection and channel estimation.In this paper,we present an approach utilizing score-based generative models to address the underdetermined nature of channel estimation,which is data-driven and well-suited for the complex and dynamic environment of massive MIMO systems.Our experimental results,based on a comprehensive dataset generated through Monte-Carlo sampling,demonstrate the high precision of our channel estimation approach,with errors reduced to as low as-45 d B,and exceptional accuracy in detecting active devices.
基金supported by the Fujian Provincial Science and Technology Program“University-Industry Cooperation Project”(2024Y4015)National Key R&D Plan of Strategic International Scientific and Technological Innovation Cooperation Project(2018YFE0207800).
文摘The increasing frequency of extreme weather events raises the likelihood of forest wildfires.Therefore,establishing an effective fire prediction model is vital for protecting human life and property,and the environment.This study aims to build a prediction model to understand the spatial characteristics and piecewise effects of forest fire drivers.Using monthly grid data from 2006 to 2020,a modeling study analyzed fire occurrences during the September to April fire season in Fujian Province,China.We compared the fitting performance of the logistic regression model(LRM),the generalized additive logistic model(GALM),and the spatial generalized additive logistic model(SGALM).The results indicate that SGALMs had the best fitting results and the highest prediction accuracy.Meteorological factors significantly impacted forest fires in Fujian Province.Areas with high fire incidence were mainly concentrated in the northwest and southeast.SGALMs improved the fitting effect of fire prediction models by considering spatial effects and the flexible fitting ability of nonlinear interpretation.This model provides piecewise interpretations of forest wildfire occurrences,which can be valuable for relevant departments and will assist forest managers in refining prevention measures based on temporal and spatial differences.
基金sponsored by the Artificial Intelligence Initiative as part of the Laboratory Directed Research and Development(LDRD)Program of Oak Ridge National Laboratory,managed by UT-Battelle,LLC,for the US Department of Energy under contract DE-AC05-00OR22725Computer resources were provided in part by the Compute and Data Environment for Science(CADES)at Oak Ridge National Laboratory.This research is also sponsored by the DOE Office of Science Award No.DESC0025591National Science Foundation Grant 2138456.
文摘Generative modeling(GM)has been increasingly used for the inverse design and optimization of materials,yet its application to molten salt mixtures remains unexplored despite how a successful approach to the inverse design of molten salts would contribute to efficiently exploiting their customizability and unlocking their advantages in applications,such as energy production and energy storage.This work presents a workflow for the inverse design of molten salts with targeted density values,addressing the challenge of representing these complex mixtures inGM.A dataset of critically evaluated molten salt densities is used to train a variational autoencoder coupled with a predictive deep neural network,which then can be used to generate new molten salt compositions with desired density values.The effectiveness of the approach is demonstrated by designing mixtures with distinct densities and validating the predicted values using ab initio molecular dynamics simulations.
基金supported by the European Union’s H2020 research and innovation program under the Marie Sklodowska-Curie grant agreement AIMed ID:861138D.P.,D.A.,G.S.and S.C.acknowledge the financial support from the European Union within the research projects 4D+nanoSCOPE ID:810316,LRI ID:C10,STOP ID:101057961+1 种基金from the German Research Foundation(DFG)within the research project UNPLOK ID:523847126,and from the“Freistaat Bayern”European Union within the project Analytiktechnikum für Gesundheits-und Umweltforschung AGEUM,StMWi-43-6623-22/1/3.
文摘Nanomaterials’properties,influenced by size,shape,and surface characteristics,are crucial for their technological,biological,and environmental applications.Accurate quantification of these materials is essential for advancing research.Deep learning segmentation networks offer precise,automated analysis,but their effectiveness depends on representative annotated datasets,which are difficult to obtain due to the high cost and manual effort required for imaging and annotation.To address this,we present DiffRenderGAN,a generative model that produces annotated synthetic data by integrating a differentiable renderer into a Generative Adversarial Network(GAN)framework.DiffRenderGAN optimizes rendering parameters to produce realistic,annotated images from non-annotated real microscopy images,reducing manual effort and improving segmentation performance compared to existing methods.Tested on ion and electron microscopy datasets,including titanium dioxide(TiO_(2)),silicon dioxide(SiO_(2)),and silver nanowires(AgNW),DiffRenderGAN bridges the gap between synthetic and real data,advancing the quantification and understanding of complex nanomaterial systems.
基金supported by the National Natural Science Foundation of China(61533019,71232006,91520301)
文摘Recently, generative adversarial networks(GANs)have become a research focus of artificial intelligence. Inspired by two-player zero-sum game, GANs comprise a generator and a discriminator, both trained under the adversarial learning idea.The goal of GANs is to estimate the potential distribution of real data samples and generate new samples from that distribution.Since their initiation, GANs have been widely studied due to their enormous prospect for applications, including image and vision computing, speech and language processing, etc. In this review paper, we summarize the state of the art of GANs and look into the future. Firstly, we survey GANs' proposal background,theoretic and implementation models, and application fields.Then, we discuss GANs' advantages and disadvantages, and their development trends. In particular, we investigate the relation between GANs and parallel intelligence,with the conclusion that GANs have a great potential in parallel systems research in terms of virtual-real interaction and integration. Clearly, GANs can provide substantial algorithmic support for parallel intelligence.
基金co-supported by the National Key Project of China(No.GJXM92579)the National Natural Science Foundation of China(Nos.92052203,61903178 and61906081)。
文摘Inverse design has long been an efficient and powerful design tool in the aircraft industry.In this paper,a novel inverse design method for supercritical airfoils is proposed based on generative models in deep learning.A Conditional Variational Auto Encoder(CVAE)and an integrated generative network CVAE-GAN that combines the CVAE with the Wasserstein Generative Adversarial Networks(WGAN),are conducted as generative models.They are used to generate target wall Mach distributions for the inverse design that matches specified features,such as locations of suction peak,shock and aft loading.Qualitative and quantitative results show that both adopted generative models can generate diverse and realistic wall Mach number distributions satisfying the given features.The CVAE-GAN model outperforms the CVAE model and achieves better reconstruction accuracies for all the samples in the dataset.Furthermore,a deep neural network for nonlinear mapping is adopted to obtain the airfoil shape corresponding to the target wall Mach number distribution.The performances of the designed deep neural network are fully demonstrated and a smoothness measurement is proposed to quantify small oscillations in the airfoil surface,proving the authenticity and accuracy of the generated airfoil shapes.
基金supported by the National Natural Science Foundation of China under Grant 51722406,52074340,and 51874335the Shandong Provincial Natural Science Foundation under Grant JQ201808+5 种基金The Fundamental Research Funds for the Central Universities under Grant 18CX02097Athe Major Scientific and Technological Projects of CNPC under Grant ZD2019-183-008the Science and Technology Support Plan for Youth Innovation of University in Shandong Province under Grant 2019KJH002the National Research Council of Science and Technology Major Project of China under Grant 2016ZX05025001-006111 Project under Grant B08028Sinopec Science and Technology Project under Grant P20050-1
文摘For reservoirs with complex non-Gaussian geological characteristics,such as carbonate reservoirs or reservoirs with sedimentary facies distribution,it is difficult to implement history matching directly,especially for the ensemble-based data assimilation methods.In this paper,we propose a multi-source information fused generative adversarial network(MSIGAN)model,which is used for parameterization of the complex geologies.In MSIGAN,various information such as facies distribution,microseismic,and inter-well connectivity,can be integrated to learn the geological features.And two major generative models in deep learning,variational autoencoder(VAE)and generative adversarial network(GAN)are combined in our model.Then the proposed MSIGAN model is integrated into the ensemble smoother with multiple data assimilation(ESMDA)method to conduct history matching.We tested the proposed method on two reservoir models with fluvial facies.The experimental results show that the proposed MSIGAN model can effectively learn the complex geological features,which can promote the accuracy of history matching.
基金The National Key R&D Program of China under contract No.2017YFE0104400the National Natural Science Foundation of China under contract No.31772852the Marine S&T Fund of Shandong Province for Pilot National Laboratory for Marine Science and Technology(Qingdao)under contract No.2018SDKJ0501-2。
文摘Habitat suitability index(HSI)models have been widely used to analyze the relationship between species abundance and environmental factors,and ultimately inform management of marine species.The response of species abundance to each environmental variable is different and habitat requirements may change over life history stages and seasons.Therefore,it is necessary to determine the optimal combination of environmental variables in HSI modelling.In this study,generalized additive models(GAMs)were used to determine which environmental variables to be included in the HSI models.Significant variables were retained and weighted in the HSI model according to their relative contribution(%)to the total deviation explained by the boosted regression tree(BRT).The HSI models were applied to evaluate the habitat suitability of mantis shrimp Oratosquilla oratoria in the Haizhou Bay and adjacent areas in 2011 and 2013–2017.Ontogenetic and seasonal variations in HSI models of mantis shrimp were also examined.Among the four models(non-optimized model,BRT informed HSI model,GAM informed HSI model,and both BRT and GAM informed HSI model),both BRT and GAM informed HSI model showed the best performance.Four environmental variables(bottom temperature,depth,distance offshore and sediment type)were selected in the HSI models for four groups(spring-juvenile,spring-adult,falljuvenile and fall-adult)of mantis shrimp.The distribution of habitat suitability showed similar patterns between juveniles and adults,but obvious seasonal variations were observed.This study suggests that the process of optimizing environmental variables in HSI models improves the performance of HSI models,and this optimization strategy could be extended to other marine organisms to enhance the understanding of the habitat suitability of target species.
基金This paper was supported by the National Natural Science Foundation of China(No.U1204606)the Key Programs for Science and Technology Development of Henan Province(No.172102210335)Key Scientific Research Projects in Henan Universities(No.16A520058).
文摘In this paper,we propose a novel coverless image steganographic scheme based on a generative model.In our scheme,the secret image is first fed to the generative model database,to generate a meaning-normal and independent image different from the secret image.The generated image is then transmitted to the receiver and fed to the generative model database to generate another image visually the same as the secret image.Thus,we only need to transmit the meaning-normal image which is not related to the secret image,and we can achieve the same effect as the transmission of the secret image.This is the first time to propose the coverless image information steganographic scheme based on generative model,compared with the traditional image steganography.The transmitted image is not embedded with any information of the secret image in this method,therefore,can effectively resist steganalysis tools.Experimental results show that our scheme has high capacity,security and reliability.
基金Project(51774219)supported by the National Natural Science Foundation of China
文摘This research develops a new mathematical modeling method by combining industrial big data and process mechanism analysis under the framework of generalized additive models(GAM)to generate a practical model with generalization and precision.Specifically,the proposed modeling method includes the following steps.Firstly,the influence factors are screened using mechanism knowledge and data-mining methods.Secondly,the unary GAM without interactions including cleaning the data,building the sub-models,and verifying the sub-models.Subsequently,the interactions between the various factors are explored,and the binary GAM with interactions is constructed.The relationships among the sub-models are analyzed,and the integrated model is built.Finally,based on the proposed modeling method,two prediction models of mechanical property and deformation resistance for hot-rolled strips are established.Industrial actual data verification demonstrates that the new models have good prediction precision,and the mean absolute percentage errors of tensile strength,yield strength and deformation resistance are 2.54%,3.34%and 6.53%,respectively.And experimental results suggest that the proposed method offers a new approach to industrial process modeling.
基金supported in part by Zhejiang Provincial Natural Science Foundation of China(LQ20F030013)Research Foundation of Hwa Mei Hospital,University of Chinese Academy of Sciences(2020HMZD22)+1 种基金Ningbo Public Service Technology Foundation(202002N3181)Medical Scientific Research Foundation of Zhejiang Province(2021431314)。
文摘Predictive models for assessing the risk of developing lung cancers can help identify high-risk individuals with the aim of recommending further screening and early intervention.To facilitate pre-hospital self-assessments,some studies have exploited predictive models trained on non-clinical data(e.g.,smoking status and family history).The performance of these models is limited due to not considering clinical data(e.g.,blood test and medical imaging results).Deep learning has shown the potential in processing complex data that combine both clinical and non-clinical information.However,predicting lung cancers remains difficult due to the severe lack of positive samples among follow-ups.To tackle this problem,this paper presents a generative-discriminative framework for improving the ability of deep learning models to generalize.According to the proposed framework,two nonlinear generative models,one based on the generative adversarial network and another on the variational autoencoder,are used to synthesize auxiliary positive samples for the training set.Then,several discriminative models,including a deep neural network(DNN),are used to assess the lung cancer risk based on a comprehensive list of risk factors.The framework was evaluated on over 55000 subjects questioned between January 2014 and December 2017,with 699 subjects being clinically diagnosed with lung cancer between January 2014 and August 2019.According to the results,the best performing predictive model built using the proposed framework was based on DNN.It achieved an average sensitivity of 76.54%and an area under the curve of 69.24%in distinguishing between the cases of lung cancer and normal cases on test sets.
基金The National Natural Science Foundation of China(Grant No.81573273,81673279,21572010 and 21772005)National Major Scientific and Technological Special Project for"Significant New Drugs Development"(Grant No.2018ZX09735001-003)
文摘Natural products（NPs） have long been recognized as a valuable resource for drug discovery, and bringing NP-related features to virtual libraries is believed to be an effective way to increase the coverage of druggable chemical space. Here, deep learning-based molecule generative model, which is a recent technique in de novo molecule design, was applied to generate virtual libraries with NP-like properties. Results demonstrated that the model was effective in generating molecules that highly resemble NPs. Moreover, the model was also found to be capable of generating NP-like molecules that were also easy to synthesize, significantly increasing the practical value of the compound library.
文摘Stochastic models are derived to estimate the level of coliform count in terms of MPN index, one of the most important water quality characteristic in ground water based on a set of water source location and soil characteristics. The study is based on about twenty location and soil characteristics, majority of them are observed through laboratory analysis of soil and water samples collected from nearly thee hundred locations of drinking water sources, wells and bore wells selected at random from the district of Kasaragod. The water contamination in wells are found to be relatively more as compared to bore wells. The study reveals that only 7 % of the wells and 40 o~ of the bore wells of the district are within the permissible limit of WHO standard of drinking water quality. The level of contamination is very high in the hospital premises and is very low in the forest area. Two separate multiple ordinal logistic regression models are developed to predict the level of coliform count, one for well and the other for bore well. The significant feature of this study is that in addition to scientifically proving the dependence of the water quality on the distances from waste disposal area and septic tanks etc., it highlights the dependence of two other very significant soil characteristics, the soil organic carbon and soil porosity. The models enable to predict the quality of water in a location based on the set of soil and location characteristics. One of the important uses of the model is in fixing safe locations for waste dump area, septic tank, digging well etc. in town planning, designing residential layouts, industrial layouts, hospital/hostel construction etc. This is the first ever study to describe the ground water quality in terms of the location and soil characteristics.
基金supported by National Natural Science Foundation of China (51774219)Science and Technology Research Program of Hubei Ministry of Education(D20161103)Youth Science and technology Program of Wuhan(2016070204010099)
文摘A model of deformation resistance during hot strip rolling was established based on generalized additive model.Firstly,a data modeling method based on generalized additive model was given.It included the selection of dependent variable and independent variables of the model,the link function of dependent variable and smoothing functional form of each independent variable,estimating process of the link function and smooth functions,and the last model modification.Then,the practical modeling test was carried out based on a large amount of hot rolling process data.An integrated variable was proposed to reflect the effects of different chemical compositions such as carbon,silicon,manganese,nickel,chromium,niobium,etc.The integrated chemical composition,strain,strain rate and rolling temperature were selected as independent variables and the cubic spline as the smooth function for them.The modeling process of deformation resistance was realized by SAS software,and the influence curves of the independent variables on deformation resistance were obtained by local scoring algorithm.Some interesting phenomena were found,for example,there is a critical value of strain rate,and the deformation resistance increases before this value and then decreases.The results confirm that the new model has higher prediction accuracy than traditional ones and is suitable for carbon steel,microalloyed steel,alloyed steel and other steel grades.
