Real-time 6 Degree-of-Freedom(DoF)pose estimation is of paramount importance for various on-orbit tasks.Benefiting from the development of deep learning,Convolutional Neural Networks(CNNs)in feature extraction has yie...Real-time 6 Degree-of-Freedom(DoF)pose estimation is of paramount importance for various on-orbit tasks.Benefiting from the development of deep learning,Convolutional Neural Networks(CNNs)in feature extraction has yielded impressive achievements for spacecraft pose estimation.To improve the robustness and interpretability of CNNs,this paper proposes a Pose Estimation approach based on Variational Auto-Encoder structure(PE-VAE)and a Feature-Aided pose estimation approach based on Variational Auto-Encoder structure(FA-VAE),which aim to accurately estimate the 6 DoF pose of a target spacecraft.Both methods treat the pose vector as latent variables,employing an encoder-decoder network with a Variational Auto-Encoder(VAE)structure.To enhance the precision of pose estimation,PE-VAE uses the VAE structure to introduce reconstruction mechanism with the whole image.Furthermore,FA-VAE enforces feature shape constraints by exclusively reconstructing the segment of the target spacecraft with the desired shape.Comparative evaluation against leading methods on public datasets reveals similar accuracy with a threefold improvement in processing speed,showcasing the significant contribution of VAE structures to accuracy enhancement,and the additional benefit of incorporating global shape prior features.展开更多
Single nucletide polymorphism(SNP)is an important factor for the study of genetic variation in human families and animal and plant strains.Therefore,it is widely used in the study of population genetics and disease re...Single nucletide polymorphism(SNP)is an important factor for the study of genetic variation in human families and animal and plant strains.Therefore,it is widely used in the study of population genetics and disease related gene.In pharmacogenomics research,identifying the association between SNP site and drug is the key to clinical precision medication,therefore,a predictive model of SNP site and drug association based on denoising variational auto-encoder(DVAE-SVM)is proposed.Firstly,k-mer algorithm is used to construct the initial SNP site feature vector,meanwhile,MACCS molecular fingerprint is introduced to generate the feature vector of the drug module.Then,we use the DVAE to extract the effective features of the initial feature vector of the SNP site.Finally,the effective feature vector of the SNP site and the feature vector of the drug module are fused input to the support vector machines(SVM)to predict the relationship of SNP site and drug module.The results of five-fold cross-validation experiments indicate that the proposed algorithm performs better than random forest(RF)and logistic regression(LR)classification.Further experiments show that compared with the feature extraction algorithms of principal component analysis(PCA),denoising auto-encoder(DAE)and variational auto-encode(VAE),the proposed algorithm has better prediction results.展开更多
The Proton Exchange Membrane Fuel Cell(PEMFC)converts the chemical energy of hydrogen fuel directly into electrical energy with broad application prospects.Understanding how current density is distributed in the PEMFC...The Proton Exchange Membrane Fuel Cell(PEMFC)converts the chemical energy of hydrogen fuel directly into electrical energy with broad application prospects.Understanding how current density is distributed in the PEMFC systems is crucial as it is a key factor influencing system performance.However,direct modeling for current distribution may encounter the challenge of dimensional catastrophe owing to the high dimensionality of the data.This paper uses a high-resolution segmented measurement device with 396 points to conduct experimental tests on the current distribution of a PEMFC with reactive area of 406 cm^(2) during a stepwise increase in load current.The current distribution is modeled based on the test results to learn the mapping relationship between the experimental parameters and the current distribution.The proposed model utilizes a Conditional Variational Auto-Encoder(CVAE)to generate current distributions.The MSE(Mean-Square Error)of the trained CVAE model reaches 9.2×10^(-5),and the comparison results show that the 222.9A current distribution error has the largest MSE of 6.36×10^(-4) and a KL Divergence(Kullback-Leibler Divergence)of 9.55×10^(-4),both of which are at a low level.This model enables the direct determination of the current distribution based on the experimental parameters,thereby establishing a technical foundation for investigating the impact of experimental conditions on fuel cells.This model is also of great significance for research on fuel cell system control strategies and fault diagnosis.展开更多
Generative AI models for music and the arts in general are increasingly complex and hard to understand.The field of ex-plainable AI(XAI)seeks to make complex and opaque AI models such as neural networks more understan...Generative AI models for music and the arts in general are increasingly complex and hard to understand.The field of ex-plainable AI(XAI)seeks to make complex and opaque AI models such as neural networks more understandable to people.One ap-proach to making generative AI models more understandable is to impose a small number of semantically meaningful attributes on gen-erative AI models.This paper contributes a systematic examination of the impact that different combinations of variational auto-en-coder models(measureVAE and adversarialVAE),configurations of latent space in the AI model(from 4 to 256 latent dimensions),and training datasets(Irish folk,Turkish folk,classical,and pop)have on music generation performance when 2 or 4 meaningful musical at-tributes are imposed on the generative model.To date,there have been no systematic comparisons of such models at this level of com-binatorial detail.Our findings show that measureVAE has better reconstruction performance than adversarialVAE which has better musical attribute independence.Results demonstrate that measureVAE was able to generate music across music genres with inter-pretable musical dimensions of control,and performs best with low complexity music such as pop and rock.We recommend that a 32 or 64 latent dimensional space is optimal for 4 regularised dimensions when using measureVAE to generate music across genres.Our res-ults are the first detailed comparisons of configurations of state-of-the-art generative AI models for music and can be used to help select and configure AI models,musical features,and datasets for more understandable generation of music.展开更多
The influenza virus changes its antigenicity frequently due to rapid mutations, leading to immune escape and failure of vaccination. Rapid determination of the influenza antigenicity could help identify the antigenic ...The influenza virus changes its antigenicity frequently due to rapid mutations, leading to immune escape and failure of vaccination. Rapid determination of the influenza antigenicity could help identify the antigenic variants in time. Here, we built a stacked auto-encoder (SAE) model for predicting the antigenic variant of human influenza A(H3N2) viruses based on the hemagglutinin (HA) protein sequences. The model achieved an accuracy of 0.95 in five-fold cross-validations, better than the logistic regression model did. Further analysis of the model shows that most of the active nodes in the hidden layer reflected the combined contribution of multiple residues to antigenic variation. Besides, some features (residues on HA protein) in the input layer were observed to take part in multiple active nodes, such as residue 189, 145 and 156, which were also reported to mostly determine the antigenic variation of influenza A(H3N2) viruses. Overall,this work is not only useful for rapidly identifying antigenic variants in influenza prevention, but also an interesting attempt in inferring the mechanisms of biological process through analysis of SAE model, which may give some insights into interpretation of the deep learning展开更多
Exposure to poor indoor air conditions poses significant risks to human health, increasing morbidity and mortality rates. Soft measurement modeling is suitable for stable and accurate monitoring of air pollutants and ...Exposure to poor indoor air conditions poses significant risks to human health, increasing morbidity and mortality rates. Soft measurement modeling is suitable for stable and accurate monitoring of air pollutants and improving air quality. Based on partial least squares (PLS), we propose an indoor air quality prediction model that utilizes variational auto-encoder regression (VAER) algorithm. To reduce the negative effects of noise, latent variables in the original data are extracted by PLS in the first step. Then, the extracted variables are used as inputs to VAER, which improve the accuracy and robustness of the model. Through comparative analysis with traditional methods, we demonstrate the superior performance of our PLS-VAER model, which exhibits improved prediction performance and stability. The root mean square error (RMSE) of PLS-VAER is reduced by 14.71%, 26.47%, and 12.50% compared to single VAER, PLS-SVR, and PLS-ANN, respectively. Additionally, the coefficient of determination (R2) of PLS-VAER improves by 13.70%, 30.09%, and 11.25% compared to single VAER, PLS-SVR, and PLS-ANN, respectively. This research offers an innovative and environmentally-friendly approach to monitor and improve indoor air quality.展开更多
In this Letter,a coding aperture design framework is introduced for data sampling of a CUP-VISAR system in laser inertial confinement fusion(ICF)research.It enhances shock wave velocity fringe reconstruction through f...In this Letter,a coding aperture design framework is introduced for data sampling of a CUP-VISAR system in laser inertial confinement fusion(ICF)research.It enhances shock wave velocity fringe reconstruction through feature fusion with a convolutional variational auto-encoder(CVAE)network.Simulation and experimental results indicate that,compared to random coding aperture,the proposed coding matrices exhibit superior reconstruction quality,achieving more accurate fringe pattern reconstruction and resolving coding information aliasing.In the experiments,the system signal-to-noise ratio(SNR)and reconstruction quality can be improved by increasing the light transmittance of the encoding matrix.This framework aids in diagnosing ICF in challenging experimental settings.展开更多
In recent years,radiotherapy based only on Magnetic Resonance(MR)images has become a hot spot for radiotherapy planning research in the current medical field.However,functional computed tomography(CT)is still needed f...In recent years,radiotherapy based only on Magnetic Resonance(MR)images has become a hot spot for radiotherapy planning research in the current medical field.However,functional computed tomography(CT)is still needed for dose calculation in the clinic.Recent deep-learning approaches to synthesized CT images from MR images have raised much research interest,making radiotherapy based only on MR images possible.In this paper,we proposed a novel unsupervised image synthesis framework with registration networks.This paper aims to enforce the constraints between the reconstructed image and the input image by registering the reconstructed image with the input image and registering the cycle-consistent image with the input image.Furthermore,this paper added ConvNeXt blocks to the network and used large kernel convolutional layers to improve the network’s ability to extract features.This research used the collected head and neck data of 180 patients with nasopharyngeal carcinoma to experiment and evaluate the training model with four evaluation metrics.At the same time,this research made a quantitative comparison of several commonly used model frameworks.We evaluate the model performance in four evaluation metrics which achieve Mean Absolute Error(MAE),Root Mean Square Error(RMSE),Peak Signal-to-Noise Ratio(PSNR),and Structural Similarity(SSIM)are 18.55±1.44,86.91±4.31,33.45±0.74 and 0.960±0.005,respectively.Compared with other methods,MAE decreased by 2.17,RMSE decreased by 7.82,PSNR increased by 0.76,and SSIM increased by 0.011.The results show that the model proposed in this paper outperforms other methods in the quality of image synthesis.The work in this paper is of guiding significance to the study of MR-only radiotherapy planning.展开更多
Single-image super-resolution(SISR)typically focuses on restoring various degraded low-resolution(LR)images to a single high-resolution(HR)image.However,during SISR tasks,it is often challenging for models to simultan...Single-image super-resolution(SISR)typically focuses on restoring various degraded low-resolution(LR)images to a single high-resolution(HR)image.However,during SISR tasks,it is often challenging for models to simultaneously maintain high quality and rapid sampling while preserving diversity in details and texture features.This challenge can lead to issues such as model collapse,lack of rich details and texture features in the reconstructed HR images,and excessive time consumption for model sampling.To address these problems,this paper proposes a Latent Feature-oriented Diffusion Probability Model(LDDPM).First,we designed a conditional encoder capable of effectively encoding LR images,reducing the solution space for model image reconstruction and thereby improving the quality of the reconstructed images.We then employed a normalized flow and multimodal adversarial training,learning from complex multimodal distributions,to model the denoising distribution.Doing so boosts the generative modeling capabilities within a minimal number of sampling steps.Experimental comparisons of our proposed model with existing SISR methods on mainstream datasets demonstrate that our model reconstructs more realistic HR images and achieves better performance on multiple evaluation metrics,providing a fresh perspective for tackling SISR tasks.展开更多
The advancements in distributed generation(DG)technologies such as solar panels have led to a widespread integration of renewable power generation in modern power systems.However,the intermittent nature of renewable e...The advancements in distributed generation(DG)technologies such as solar panels have led to a widespread integration of renewable power generation in modern power systems.However,the intermittent nature of renewable energy poses new challenges to the network operational planning with underlying uncertainties.This paper proposes a novel probabilistic scheme for renewable solar power generation forecasting by addressing data and model parameter uncertainties using Bayesian bidirectional long short-term memory(BiLSTM)neural networks,while handling the high dimensionality in weight parameters using variational auto-encoders(VAE).The forecasting performance of the proposed method is evaluated using various deterministic and probabilistic evaluation metrics such as root-mean square error(RMSE),Pinball loss,etc.Furthermore,reconstruction error and computational time are also monitored to evaluate the dimensionality reduction using the VAE component.When compared with benchmark methods,the proposed method leads to significant improvements in weight reduction,i.e.,from 76,4224 to 2,022 number of weight parameters,quantifying to 97.35%improvement in weight parameters reduction and 37.93%improvement in computational time for 6 months of solar power generation data.展开更多
Word-embedding acts as one of the backbones of modern natural language processing(NLP).Recently,with the need for deploying NLP models to low-resource devices,there has been a surge of interest to compress word embedd...Word-embedding acts as one of the backbones of modern natural language processing(NLP).Recently,with the need for deploying NLP models to low-resource devices,there has been a surge of interest to compress word embeddings into hash codes or binary vectors so as to save the storage and memory consumption.Typically,existing work learns to encode an embedding into a compressed representation from which the original embedding can be reconstructed.Although these methods aim to preserve most information of every individual word,they often fail to retain the relation between words,thus can yield large loss on certain tasks.To this end,this paper presents Relation Reconstructive Binarization(R2B)to transform word embeddings into binary codes that can preserve the relation between words.At its heart,R2B trains an auto-encoder to generate binary codes that allow reconstructing the wordby-word relations in the original embedding space.Experiments showed that our method achieved significant improvements over previous methods on a number of tasks along with a space-saving of up to 98.4%.Specifically,our method reached even better results on word similarity evaluation than the uncompressed pre-trained embeddings,and was significantly better than previous compression methods that do not consider word relations.展开更多
基金supported by the National Natural Science Foundation of China(No.52272390)the Natural Science Foundation of Heilongjiang Province of China(No.YQ2022A009)the Shanghai Sailing Program,China(No.20YF1417300).
文摘Real-time 6 Degree-of-Freedom(DoF)pose estimation is of paramount importance for various on-orbit tasks.Benefiting from the development of deep learning,Convolutional Neural Networks(CNNs)in feature extraction has yielded impressive achievements for spacecraft pose estimation.To improve the robustness and interpretability of CNNs,this paper proposes a Pose Estimation approach based on Variational Auto-Encoder structure(PE-VAE)and a Feature-Aided pose estimation approach based on Variational Auto-Encoder structure(FA-VAE),which aim to accurately estimate the 6 DoF pose of a target spacecraft.Both methods treat the pose vector as latent variables,employing an encoder-decoder network with a Variational Auto-Encoder(VAE)structure.To enhance the precision of pose estimation,PE-VAE uses the VAE structure to introduce reconstruction mechanism with the whole image.Furthermore,FA-VAE enforces feature shape constraints by exclusively reconstructing the segment of the target spacecraft with the desired shape.Comparative evaluation against leading methods on public datasets reveals similar accuracy with a threefold improvement in processing speed,showcasing the significant contribution of VAE structures to accuracy enhancement,and the additional benefit of incorporating global shape prior features.
基金Lanzhou Talent Innovation and Entrepreneurship Project(No.2020-RC-14)。
文摘Single nucletide polymorphism(SNP)is an important factor for the study of genetic variation in human families and animal and plant strains.Therefore,it is widely used in the study of population genetics and disease related gene.In pharmacogenomics research,identifying the association between SNP site and drug is the key to clinical precision medication,therefore,a predictive model of SNP site and drug association based on denoising variational auto-encoder(DVAE-SVM)is proposed.Firstly,k-mer algorithm is used to construct the initial SNP site feature vector,meanwhile,MACCS molecular fingerprint is introduced to generate the feature vector of the drug module.Then,we use the DVAE to extract the effective features of the initial feature vector of the SNP site.Finally,the effective feature vector of the SNP site and the feature vector of the drug module are fused input to the support vector machines(SVM)to predict the relationship of SNP site and drug module.The results of five-fold cross-validation experiments indicate that the proposed algorithm performs better than random forest(RF)and logistic regression(LR)classification.Further experiments show that compared with the feature extraction algorithms of principal component analysis(PCA),denoising auto-encoder(DAE)and variational auto-encode(VAE),the proposed algorithm has better prediction results.
基金sponsored by Science and Technology Program of Sichuan Province(2024ZDZX0035 and 2024ZHCG0072)。
文摘The Proton Exchange Membrane Fuel Cell(PEMFC)converts the chemical energy of hydrogen fuel directly into electrical energy with broad application prospects.Understanding how current density is distributed in the PEMFC systems is crucial as it is a key factor influencing system performance.However,direct modeling for current distribution may encounter the challenge of dimensional catastrophe owing to the high dimensionality of the data.This paper uses a high-resolution segmented measurement device with 396 points to conduct experimental tests on the current distribution of a PEMFC with reactive area of 406 cm^(2) during a stepwise increase in load current.The current distribution is modeled based on the test results to learn the mapping relationship between the experimental parameters and the current distribution.The proposed model utilizes a Conditional Variational Auto-Encoder(CVAE)to generate current distributions.The MSE(Mean-Square Error)of the trained CVAE model reaches 9.2×10^(-5),and the comparison results show that the 222.9A current distribution error has the largest MSE of 6.36×10^(-4) and a KL Divergence(Kullback-Leibler Divergence)of 9.55×10^(-4),both of which are at a low level.This model enables the direct determination of the current distribution based on the experimental parameters,thereby establishing a technical foundation for investigating the impact of experimental conditions on fuel cells.This model is also of great significance for research on fuel cell system control strategies and fault diagnosis.
文摘Generative AI models for music and the arts in general are increasingly complex and hard to understand.The field of ex-plainable AI(XAI)seeks to make complex and opaque AI models such as neural networks more understandable to people.One ap-proach to making generative AI models more understandable is to impose a small number of semantically meaningful attributes on gen-erative AI models.This paper contributes a systematic examination of the impact that different combinations of variational auto-en-coder models(measureVAE and adversarialVAE),configurations of latent space in the AI model(from 4 to 256 latent dimensions),and training datasets(Irish folk,Turkish folk,classical,and pop)have on music generation performance when 2 or 4 meaningful musical at-tributes are imposed on the generative model.To date,there have been no systematic comparisons of such models at this level of com-binatorial detail.Our findings show that measureVAE has better reconstruction performance than adversarialVAE which has better musical attribute independence.Results demonstrate that measureVAE was able to generate music across music genres with inter-pretable musical dimensions of control,and performs best with low complexity music such as pop and rock.We recommend that a 32 or 64 latent dimensional space is optimal for 4 regularised dimensions when using measureVAE to generate music across genres.Our res-ults are the first detailed comparisons of configurations of state-of-the-art generative AI models for music and can be used to help select and configure AI models,musical features,and datasets for more understandable generation of music.
文摘The influenza virus changes its antigenicity frequently due to rapid mutations, leading to immune escape and failure of vaccination. Rapid determination of the influenza antigenicity could help identify the antigenic variants in time. Here, we built a stacked auto-encoder (SAE) model for predicting the antigenic variant of human influenza A(H3N2) viruses based on the hemagglutinin (HA) protein sequences. The model achieved an accuracy of 0.95 in five-fold cross-validations, better than the logistic regression model did. Further analysis of the model shows that most of the active nodes in the hidden layer reflected the combined contribution of multiple residues to antigenic variation. Besides, some features (residues on HA protein) in the input layer were observed to take part in multiple active nodes, such as residue 189, 145 and 156, which were also reported to mostly determine the antigenic variation of influenza A(H3N2) viruses. Overall,this work is not only useful for rapidly identifying antigenic variants in influenza prevention, but also an interesting attempt in inferring the mechanisms of biological process through analysis of SAE model, which may give some insights into interpretation of the deep learning
基金supported by the Opening Project of Guangxi Key Laboratory of Clean Pulp&Papermaking and Pollution Control,China(No.2021KF11)the Shandong Provincial Natural Science Foundation,China(No.ZR2021MF135)+1 种基金the National Natural Science Foundation of China(No.52170001)the Natural Science Foundation of Jiangsu Provincial Universities,China(No.22KJA530003).
文摘Exposure to poor indoor air conditions poses significant risks to human health, increasing morbidity and mortality rates. Soft measurement modeling is suitable for stable and accurate monitoring of air pollutants and improving air quality. Based on partial least squares (PLS), we propose an indoor air quality prediction model that utilizes variational auto-encoder regression (VAER) algorithm. To reduce the negative effects of noise, latent variables in the original data are extracted by PLS in the first step. Then, the extracted variables are used as inputs to VAER, which improve the accuracy and robustness of the model. Through comparative analysis with traditional methods, we demonstrate the superior performance of our PLS-VAER model, which exhibits improved prediction performance and stability. The root mean square error (RMSE) of PLS-VAER is reduced by 14.71%, 26.47%, and 12.50% compared to single VAER, PLS-SVR, and PLS-ANN, respectively. Additionally, the coefficient of determination (R2) of PLS-VAER improves by 13.70%, 30.09%, and 11.25% compared to single VAER, PLS-SVR, and PLS-ANN, respectively. This research offers an innovative and environmentally-friendly approach to monitor and improve indoor air quality.
基金supported by the National Natural Science Foundation of China(Nos.12127810 and 61604028)the Science and Technology Research Program of Chongqing Municipal Education Commission(No.KJZD-K202200607)。
文摘In this Letter,a coding aperture design framework is introduced for data sampling of a CUP-VISAR system in laser inertial confinement fusion(ICF)research.It enhances shock wave velocity fringe reconstruction through feature fusion with a convolutional variational auto-encoder(CVAE)network.Simulation and experimental results indicate that,compared to random coding aperture,the proposed coding matrices exhibit superior reconstruction quality,achieving more accurate fringe pattern reconstruction and resolving coding information aliasing.In the experiments,the system signal-to-noise ratio(SNR)and reconstruction quality can be improved by increasing the light transmittance of the encoding matrix.This framework aids in diagnosing ICF in challenging experimental settings.
基金supported by the National Science Foundation for Young Scientists of China(Grant No.61806060)2019-2021,the Basic and Applied Basic Research Foundation of Guangdong Province(2021A1515220140)the Youth Innovation Project of Sun Yat-sen University Cancer Center(QNYCPY32).
文摘In recent years,radiotherapy based only on Magnetic Resonance(MR)images has become a hot spot for radiotherapy planning research in the current medical field.However,functional computed tomography(CT)is still needed for dose calculation in the clinic.Recent deep-learning approaches to synthesized CT images from MR images have raised much research interest,making radiotherapy based only on MR images possible.In this paper,we proposed a novel unsupervised image synthesis framework with registration networks.This paper aims to enforce the constraints between the reconstructed image and the input image by registering the reconstructed image with the input image and registering the cycle-consistent image with the input image.Furthermore,this paper added ConvNeXt blocks to the network and used large kernel convolutional layers to improve the network’s ability to extract features.This research used the collected head and neck data of 180 patients with nasopharyngeal carcinoma to experiment and evaluate the training model with four evaluation metrics.At the same time,this research made a quantitative comparison of several commonly used model frameworks.We evaluate the model performance in four evaluation metrics which achieve Mean Absolute Error(MAE),Root Mean Square Error(RMSE),Peak Signal-to-Noise Ratio(PSNR),and Structural Similarity(SSIM)are 18.55±1.44,86.91±4.31,33.45±0.74 and 0.960±0.005,respectively.Compared with other methods,MAE decreased by 2.17,RMSE decreased by 7.82,PSNR increased by 0.76,and SSIM increased by 0.011.The results show that the model proposed in this paper outperforms other methods in the quality of image synthesis.The work in this paper is of guiding significance to the study of MR-only radiotherapy planning.
基金supported by General Project of Guangxi Science and Technology Major Project(AA19254016)Beihai City Science and Technology Planning Project(202082033)+1 种基金Beihai City Science and Technology Planning Project(202082023)Guangxi Graduate Student Innovation Project(YCSW2021174)。
文摘Single-image super-resolution(SISR)typically focuses on restoring various degraded low-resolution(LR)images to a single high-resolution(HR)image.However,during SISR tasks,it is often challenging for models to simultaneously maintain high quality and rapid sampling while preserving diversity in details and texture features.This challenge can lead to issues such as model collapse,lack of rich details and texture features in the reconstructed HR images,and excessive time consumption for model sampling.To address these problems,this paper proposes a Latent Feature-oriented Diffusion Probability Model(LDDPM).First,we designed a conditional encoder capable of effectively encoding LR images,reducing the solution space for model image reconstruction and thereby improving the quality of the reconstructed images.We then employed a normalized flow and multimodal adversarial training,learning from complex multimodal distributions,to model the denoising distribution.Doing so boosts the generative modeling capabilities within a minimal number of sampling steps.Experimental comparisons of our proposed model with existing SISR methods on mainstream datasets demonstrate that our model reconstructs more realistic HR images and achieves better performance on multiple evaluation metrics,providing a fresh perspective for tackling SISR tasks.
文摘The advancements in distributed generation(DG)technologies such as solar panels have led to a widespread integration of renewable power generation in modern power systems.However,the intermittent nature of renewable energy poses new challenges to the network operational planning with underlying uncertainties.This paper proposes a novel probabilistic scheme for renewable solar power generation forecasting by addressing data and model parameter uncertainties using Bayesian bidirectional long short-term memory(BiLSTM)neural networks,while handling the high dimensionality in weight parameters using variational auto-encoders(VAE).The forecasting performance of the proposed method is evaluated using various deterministic and probabilistic evaluation metrics such as root-mean square error(RMSE),Pinball loss,etc.Furthermore,reconstruction error and computational time are also monitored to evaluate the dimensionality reduction using the VAE component.When compared with benchmark methods,the proposed method leads to significant improvements in weight reduction,i.e.,from 76,4224 to 2,022 number of weight parameters,quantifying to 97.35%improvement in weight parameters reduction and 37.93%improvement in computational time for 6 months of solar power generation data.
基金The reseach work was supported by the National Key Research and Development Program of China(2017YFB1002104)the National Natural Science Foundation of China(Grant Nos.92046003,61976204,U1811461)Xiang Ao was also supported by the Project of Youth Innovation Promotion Association CAS and Beijing Nova Program(Z201100006820062).
文摘Word-embedding acts as one of the backbones of modern natural language processing(NLP).Recently,with the need for deploying NLP models to low-resource devices,there has been a surge of interest to compress word embeddings into hash codes or binary vectors so as to save the storage and memory consumption.Typically,existing work learns to encode an embedding into a compressed representation from which the original embedding can be reconstructed.Although these methods aim to preserve most information of every individual word,they often fail to retain the relation between words,thus can yield large loss on certain tasks.To this end,this paper presents Relation Reconstructive Binarization(R2B)to transform word embeddings into binary codes that can preserve the relation between words.At its heart,R2B trains an auto-encoder to generate binary codes that allow reconstructing the wordby-word relations in the original embedding space.Experiments showed that our method achieved significant improvements over previous methods on a number of tasks along with a space-saving of up to 98.4%.Specifically,our method reached even better results on word similarity evaluation than the uncompressed pre-trained embeddings,and was significantly better than previous compression methods that do not consider word relations.
