As a mature distributed machine learning paradigm,federated learning enables wireless edge devices to collaboratively train a shared AI-model by stochastic gradient descent(SGD).However,devices need to upload high-dim...As a mature distributed machine learning paradigm,federated learning enables wireless edge devices to collaboratively train a shared AI-model by stochastic gradient descent(SGD).However,devices need to upload high-dimensional stochastic gradients to edge server in training,which cause severe communication bottleneck.To address this problem,we compress the communication by sparsifying and quantizing the stochastic gradients of edge devices.We first derive a closed form of the communication compression in terms of sparsification and quantization factors.Then,the convergence rate of this communicationcompressed system is analyzed and several insights are obtained.Finally,we formulate and deal with the quantization resource allocation problem for the goal of minimizing the convergence upper bound,under the constraint of multiple-access channel capacity.Simulations show that the proposed scheme outperforms the benchmarks.展开更多
Submarine groundwater discharge(SGD),which can be traced using naturally occurring radium isotopes,has been recognized as a significant nutrient source and land-ocean interaction passage for the coastal waters of the ...Submarine groundwater discharge(SGD),which can be traced using naturally occurring radium isotopes,has been recognized as a significant nutrient source and land-ocean interaction passage for the coastal waters of the Daya Bay,China.However,uncertainties in assessing SGD fluxes must still be discussed in detail.In this study,we attempted to utilize the Monte Carlo method to evaluate the uncertainties of radium-derived SGD flux in the northeast and entirety of the Daya Bay.The results show that the uncertainties of the SGD estimate in the northeast bay are very sensitive to variations in excess radium inventories as well as radium inputs from bottom sediments,while the uncertainties of the SGD estimate for the entire bay are strongly affected by fluctuations in radium inputs from bottom sediments and radium end-members of SGD.This study will help to distinguish the key factors controlling the accuracy of SGD estimates in similar coastal waters.展开更多
Deep learning has reached many successes in Video Processing.Video has become a growing important part of our daily digital interactions.The advancement of better resolution content and the large volume offers serious...Deep learning has reached many successes in Video Processing.Video has become a growing important part of our daily digital interactions.The advancement of better resolution content and the large volume offers serious challenges to the goal of receiving,distributing,compressing and revealing highquality video content.In this paper we propose a novel Effective and Efficient video compression by the Deep Learning framework based on the flask,which creatively combines the Deep Learning Techniques on Convolutional Neural Networks(CNN)and Generative Adversarial Networks(GAN).The video compression method involves the layers are divided into different groups for data processing,using CNN to remove the duplicate frames,repeating the single image instead of the duplicate images by recognizing and detecting minute changes using GAN and recorded with Long Short-Term Memory(LSTM).Instead of the complete image,the small changes generated using GAN are substituted,which helps with frame-level compression.Pixel wise comparison is performed using K-nearest Neighbours(KNN)over the frame,clustered with K-means and Singular Value Decomposition(SVD)is applied for every frame in the video for all three colour channels[Red,Green,Blue]to decrease the dimension of the utility matrix[R,G,B]by extracting its latent factors.Video frames are packed with parameters with the aid of a codec and converted to video format and the results are compared with the original video.Repeated experiments on several videos with different sizes,duration,Frames per second(FPS),and quality results demonstrated a significant resampling rate.On normal,the outcome delivered had around a 10%deviation in quality and over half in size when contrasted,and the original video.展开更多
提出了SGD(shape global deformation)算法并将其应用于回弹补偿分析中,进行补偿面的修改。其原理是以原始曲面为基础进行的曲面整体变形修改,这种方法能够保证变形前后的型面整体拓扑关系的一致性,保证曲面的质量、光顺程度,以便获得...提出了SGD(shape global deformation)算法并将其应用于回弹补偿分析中,进行补偿面的修改。其原理是以原始曲面为基础进行的曲面整体变形修改,这种方法能够保证变形前后的型面整体拓扑关系的一致性,保证曲面的质量、光顺程度,以便获得更精确的回弹补偿型面。经多个实例分析表明,该算法能够在很大程度上节省修改补偿型面的时间和精力,较大地提高曲面精度,回弹补偿更加准确。展开更多
基金supported in part by the National Key Research and Development Program of China under Grant 2020YFB1807700in part by the National Science Foundation of China under Grant U200120122
文摘As a mature distributed machine learning paradigm,federated learning enables wireless edge devices to collaboratively train a shared AI-model by stochastic gradient descent(SGD).However,devices need to upload high-dimensional stochastic gradients to edge server in training,which cause severe communication bottleneck.To address this problem,we compress the communication by sparsifying and quantizing the stochastic gradients of edge devices.We first derive a closed form of the communication compression in terms of sparsification and quantization factors.Then,the convergence rate of this communicationcompressed system is analyzed and several insights are obtained.Finally,we formulate and deal with the quantization resource allocation problem for the goal of minimizing the convergence upper bound,under the constraint of multiple-access channel capacity.Simulations show that the proposed scheme outperforms the benchmarks.
基金The Project of Key Laboratory of Marine Environmental Survey Technology and Application,Ministry of Natural Resources under contract No.MESTA-2021-D006the China Ocean Development Foundation under contract No.CODF-002-ZX-2021+5 种基金the Science and Technology Plan Projects of Guangdong Province under contract No.2021B1212050025the Director’s Foundation of South China Sea Bureau of Ministry of Natural Resources under contract No.230201the Research Fund Program of Guangdong Provincial Key Laboratory of Applied Marine Biology under contract No.LAMB20221007the Natural Science Foundation of Guangdong Province of China under contract No.2017A030310592the Key Program of Bureau Director of State Oceanic Administration under contract No.180104the Open Project of State Key Laboratory of Tropical Oceanography,South China Sea Institute of Oceanology,Chinese Academy of Sciences under contract No.LTO1709.
文摘Submarine groundwater discharge(SGD),which can be traced using naturally occurring radium isotopes,has been recognized as a significant nutrient source and land-ocean interaction passage for the coastal waters of the Daya Bay,China.However,uncertainties in assessing SGD fluxes must still be discussed in detail.In this study,we attempted to utilize the Monte Carlo method to evaluate the uncertainties of radium-derived SGD flux in the northeast and entirety of the Daya Bay.The results show that the uncertainties of the SGD estimate in the northeast bay are very sensitive to variations in excess radium inventories as well as radium inputs from bottom sediments,while the uncertainties of the SGD estimate for the entire bay are strongly affected by fluctuations in radium inputs from bottom sediments and radium end-members of SGD.This study will help to distinguish the key factors controlling the accuracy of SGD estimates in similar coastal waters.
文摘Deep learning has reached many successes in Video Processing.Video has become a growing important part of our daily digital interactions.The advancement of better resolution content and the large volume offers serious challenges to the goal of receiving,distributing,compressing and revealing highquality video content.In this paper we propose a novel Effective and Efficient video compression by the Deep Learning framework based on the flask,which creatively combines the Deep Learning Techniques on Convolutional Neural Networks(CNN)and Generative Adversarial Networks(GAN).The video compression method involves the layers are divided into different groups for data processing,using CNN to remove the duplicate frames,repeating the single image instead of the duplicate images by recognizing and detecting minute changes using GAN and recorded with Long Short-Term Memory(LSTM).Instead of the complete image,the small changes generated using GAN are substituted,which helps with frame-level compression.Pixel wise comparison is performed using K-nearest Neighbours(KNN)over the frame,clustered with K-means and Singular Value Decomposition(SVD)is applied for every frame in the video for all three colour channels[Red,Green,Blue]to decrease the dimension of the utility matrix[R,G,B]by extracting its latent factors.Video frames are packed with parameters with the aid of a codec and converted to video format and the results are compared with the original video.Repeated experiments on several videos with different sizes,duration,Frames per second(FPS),and quality results demonstrated a significant resampling rate.On normal,the outcome delivered had around a 10%deviation in quality and over half in size when contrasted,and the original video.
文摘提出了SGD(shape global deformation)算法并将其应用于回弹补偿分析中,进行补偿面的修改。其原理是以原始曲面为基础进行的曲面整体变形修改,这种方法能够保证变形前后的型面整体拓扑关系的一致性,保证曲面的质量、光顺程度,以便获得更精确的回弹补偿型面。经多个实例分析表明,该算法能够在很大程度上节省修改补偿型面的时间和精力,较大地提高曲面精度,回弹补偿更加准确。
