|
|
|
题名
|
作者
|
年代
|
出处
|
被引量
|
| 1 | DropConnect Regularization Method with Sparsity Constraint for Neural Networks显示文摘Drop Connect is a recently introduced algorithm to prevent the co-adaptation of feature detectors.Compared to Dropout, Drop Connect gains state-of-the-art results on several image recognition benchmarks. Motivated by the success of Drop Connect, we extended this algorithm with the ability of sparse feature selection. In Drop Connect algorithm, the dropping masks of weights are generated using Bernoulli gating variables that are independent of the weights and activations. We introduce a new strategy to generate masks depending on the outputs of previous layer. Using this method, neurons which are promising to produce sparser features will be assigned a bigger possibility to keep active in the forward and backward propagations. We then evaluate such sparsity constrained Drop Connect on MNIST and CIFAR datasets in comparison with ordinary Drop Connect and Dropout method. The results show that our new method improves the sparsity of features significantly, while not degrading the precision. | LIAN Zifeng JING Xiaojun WANG Xiaohan HUANG Hai TAN Youheng CUI Yuanhao | 2016 | Chinese Journal of Electronics2016，25，1： | 6 |
| 2 | A vacuum ultraviolet laser with a submicrometer spot for spatially resolved photoemission spectroscopy显示文摘Vacuum ultraviolet(VUV)lasers have demonstrated great potential as the light source for various spectroscopies,which,if they can be focused into a small beam spot,will not only allow investigation of mesoscopic materials and structures but also find application in the manufacture of nano-objects with excellent precision.In this work,we report the construction of a 177 nm VUV laser that can achieve a record-small(〜0.76 μm)focal spot at a long focal length(~45 mm)by using a flat lens without spherical aberration.The size of the beam spot of this VUV laser was tested using a metal grating and exfoliated graphene flakes,and we demonstrated its application in a fluorescence spectroscopy study on pure and Tm3+-doped NaYF4 microcrystals,revealing a new emission band that cannot be observed in the traditional up-conversion process.In addition,this laser system would be an ideal light source for spatially and angleresolved photoemission spectroscopy. | Yuanhao Mao Dong Zhao Shen Yan Hongjia Zhang Juan Li Kai Han Xiaojun Xu Chuan Guo Lexian Yang Chaofan Zhang Kun Huang Yulin Chen | 2021 | Light(Science & Applications)2021，10，3： | 3 |
| 3 | Energy - Sav- ing Cooperative Spectrum Sensing Processor for Cognitive Radio System 显示文摘 | CHEN Wenbin YANG Chikai HUANG Yuanhao | 2011 | Circuits and Systems I: Reqular Papers IEEE2011，58，4： | 1 |
| 4 | Energy Saving Cooperative Spectrum Sensing Processor for Cognitive Radio System 显示文摘 | Chen Wenbin Yang Chikai Huang Yuanhao | 2011 | IEEE Transactions on Regular Papers2011，58，4： | 1 |
| 5 | Energy Saving Cooperative Spectrum Sensing Processor for Cognitive Radio System显示文摘 | Chen Wenbin Yang Chikai Huang Yuanhao | 2011 | IEEE Transactions on Circuits and Systems I:Reqular Papers2011，58，4： | 1 |
| 6 | A new audio coding scheme using a forward masking model and perceptually weighted vector quantization 显示文摘 | HUANG Yuanhao Chiueh T D | 2002 | IEEE Trans Speech and Audio Processing2002，10，5： | 1 |
| 7 | Spintronics intelligent devices显示文摘Intelligent computing paradigms have become increasingly important for the efficient processing of massive amounts of data.However,using traditional electronic devices to implement these intelligent paradigms is currently mismatched and limited by their energy,area,and speed.Spintronics,which exploits the magnetic and electrical properties of electrons,could break through these limitations and bring new possibilities to electrical devices.In particular,the tunneling magnetoresistance effect,merging quantum and spintronics,enables spintronic devices to be compatible with standard integrated circuits with a magnetic tunnel junction(MTJ)design,showing great potential for implementing hardware-based intelligent frameworks.In this review,we introduce the specific capabilities of MTJs,including nonvolatility,stochasticity,plasticity,and nonlinearity,which are highly favorable in artificial intelligence algorithms.We then present how these devices could impact the development of intelligent computing,including in-memory computing,probabilistic computing,and neuromorphic computing.Finally,we discuss their challenges and perspectives in intelligent hardware implementations. | Wenlong Cai Yan Huang Xueying Zhang Shihong Wang Yuanhao Pan Jialiang Yin Kewen Shi Weisheng Zhao | 2023 | Science China(Physics,Mechanics & Astronomy)2023，66，11： | 0 |
| 8 | Polydopamine nanolayer assisted internal photo-deposition of CdS nanocrystals for stable cosensitized photoanode显示文摘Hydrogen evolution via photo-electro-chemical(PEC)co-catalysis is potential for solving energy crisis and environmental issues.The rapidly advances of fabrication and broad applications of polydopamine(PDA)and its derivatives have drawn intense attentions in recent years.Herein,an ultrathin PDA coating with nanometer accuracy was conformally grown on TiO_(2) nanotube arrays(NTAs)via electrochemical polymerization,in which the polymer provided a platform for further photoinduced assembly of CdS nanocrystals in the embedded mode.The optimized CdS@PDA/TiO_(2) NTAs hierarchical heterostructure as photoanode gave an excellent PEC performance and exhibited outstanding stability under light irradiation.The photocurrent density was heightened to 5.48 mA·cm^(–2),which was beneficial to H_(2) evolution with a rate of 20μmol·h^(–1)·cm^(–2).The improvement of PEC activity was ascribed to co-photosensitization,optimized carriers transfer,and transport route arised from CdS embedding,resulting to provide a persistent driving force for charge separation based on secure heterojunction of CdS/TiO_(2) glued by PDA.The improvement of PEC stability was due to the inhibition of CdS photocorrosion covered by PDA shelter.This advance boded well for the development of PEC field founded on multifunctional PDA. | Na Li Lu Han Hainan Zhang Juntong Huang Xudong Luo Xibao Li Yuanhao Wang Weiqi Qian Ya Yang | 2022 | Nano Research2022，15，10： | 0 |
| 9 | Controllable image generation based on causal representation learning显示文摘Artificial intelligence generated content(AIGC)has emerged as an indispensable tool for producing large-scale content in various forms,such as images,thanks to the significant role that AI plays in imitation and production.However,interpretability and controllability remain challenges.Existing AI methods often face challenges in producing images that are both flexible and controllable while considering causal relationships within the images.To address this issue,we have developed a novel method for causal controllable image generation(CCIG)that combines causal representation learning with bi-directional generative adversarial networks(GANs).This approach enables humans to control image attributes while considering the rationality and interpretability of the generated images and also allows for the generation of counterfactual images.The key of our approach,CCIG,lies in the use of a causal structure learning module to learn the causal relationships between image attributes and joint optimization with the encoder,generator,and joint discriminator in the image generation module.By doing so,we can learn causal representations in image’s latent space and use causal intervention operations to control image generation.We conduct extensive experiments on a real-world dataset,CelebA.The experimental results illustrate the effectiveness of CCIG. | Shanshan HUANG Yuanhao WANG Zhili GONG Jun LIAO Shu WANG Li LIU | 2024 | Frontiers of Information Technology & Electronic Engineering2024，25，1： | 0 |
| 10 | Genetic Insights into Glycine’s Protective Role Against CAD—European and East Asia,2015 and 2020显示文摘Introduction:The purpose of this study is to examine the potential causal relationship between levels of circulating glycine and coronary artery disease(CAD)using a two-step Mendelian randomization(MR)analysis.Methods:We analyzed data from genome-wide association studies(GWAS)conducted on European and East Asian populations.To assess the causal effects of circulating glycine levels on the risk of CAD.We used the inverse-variance weighting(IVW),weighted median(WM),MR-Egger,and Mendelian Randomization Pleiotropy RESidual Sum and Outlier(MR-PRESSO)methods.Furthermore,we conducted mediation analysis to investigate the contribution of blood pressure and other cardiovascular disease-related traits.Results:The two-step Mendelian randomization analysis revealed that higher levels of glycine in the blood were associated with a reduced risk of CAD in Europeans[odds ratio(OR)=0.84,95%confidence interval(CI):0.72,−0.98;P=0.029]and East Asians:(OR=0.76,95%CI:0.66,−0.89;P=3.57×10−4).Sensitivity analysis confirmed the robustness of these findings.Additionally,our results suggest that about 6.06%of the observed causal effect is mediated through genetically predicted systolic blood pressure(SBP)in the European population.Discussion:Our results contribute to the current knowledge regarding the involvement of glycine in the progression of CAD,and provide valuable methodological insights for the prevention and treatment of this condition. | Jiaying Yu Zhuolin Zhu Ting Wang Yuanhao Wei Lianjie Huang Qianru Zhang Yuting Zhang Yiran Wang Guiyou Liu Xiang Shu Rennan Feng | 2024 | China CDC weekly2024，6，9： | 0 |
| 11 | Joint view synthesis and disparity refinement for stereo matching显示文摘Typical stereo algorithms treat disparity estimation and view synthesis as two sequential procedures.In this paper,we consider stereo matching and view synthesis as two complementary components,and present a novel iterative refinement model for joint view synthesis and disparity refinement.To achieve the mutual promotion between view synthesis and disparity refinement,we apply two key strategies,disparity maps fusion and disparity-assisted plane sweep-based rendering(DAPSR).On the one hand,the disparity maps fusion strategy is applied to generate disparity map from synthesized view and input views.This strategy is able to detect and counteract disparity errors caused by potential artifacts from synthesized view.On the other hand,the DAPSR is used for view synthesis and updating,and is able to weaken the interpolation errors caused by outliers in the disparity maps.Experiments on Middlebury benchmarks demonstrate that by introducing the synthesized view,disparity errors due to large occluded region and large baseline are eliminated effectively and the synthesis quality is greatly improved. | Gaochang WU Yipeng LI Yuanhao HUANG Yebin LIU | 2019 | Frontiers of Computer Science2019，13，6： | 0 |
| 12 | The light and hypoxia induced gene ZmPORB1 determines tocopherol content in the maize kernel显示文摘Tocopherol is an important lipid-soluble antioxidant beneficial for both human health and plant growth. Here, we fine mapped a major QTLqVE1 affecting γ-tocopherol content in maize kernel, positionally cloned and confirmed the underlying gene ZmPORB1(por1), as a protochlorophyllide oxidoreductase. A 13.7 kb insertion reduced the tocopherol and chlorophyll content, and the photosynthetic activity by repressing ZmPORB1 expression in embryos of NIL-K22, but did not affect the levels of the tocopherol precursors HGA(homogentisic acid)and PMP(phytyl monophosphate). Furthermore, ZmPORB1 is inducible by low oxygen and light, thereby involved in the hypoxia response in developing embryos. Concurrent with natural hypoxia in embryos, the redox state has been changed with NO increasing and H_(2)O_(2) decreasing, which lowered γ-tocopherol content via scavenging reactive nitrogen species. In conclusion, we proposed that the lower lightharvesting chlorophyll content weakened embryo photosynthesis, leading to fewer oxygen supplies and consequently diverse hypoxic responses including an elevated γ-tocopherol consumption. Our findings shed light on the mechanism for fine-tuning endogenous oxygen concentration in the maize embryo through a novel feedback pathway involving the light and low oxygen regulation of ZmPORB1 expression and chlorophyll content. | Nannan Liu Yuanhao Du Shijuan Yan Wei Chen Min Deng Shutu Xu Hong Wang Wei Zhan Wenjie Huang Yan Yin Xiaohong Yang Qiao Zhao Alisdair R.Fernie Jianbing Yan | 2024 | Science China(Life Sciences)2024，67，3： | 0 |
| 13 | Numerical differentiation of noisy data with local optimum by data segmentation显示文摘A new numerical differentiation method with local optimum by data segmentation is proposed. The segmentation of data is based on the second derivatives computed by a Fourier development method. A filtering process is used to achieve acceptable segmentation. Numerical results are presented by using the data segmentation method, compared with the regularization method.For further investigation, the proposed algorithm is applied to the resistance capacitance(RC) networks identification problem, and improvements of the result are obtained by using this algorithm. | Jianhua Zhang Xiufu Que Wei Chen Yuanhao Huang Lianqiao Yang | 2015 | Journal of Systems Engineering and Electronics2015，26，4： | 0 |
| 14 | Identification of Suitable Reference Genes for qRT-PCR Normalization in Tilia miqueliana Maxim显示文摘Quantitative real-time polymerase chain reaction(qRT-PCR)is a rapid and effective approach toward detecting the expression patterns of target genes.The selection of a stable reference gene under specific test condition is essential for expressing levels of target genes accurately.Tilia miqueliana,considered endangered,is a prominent native ornamental and honey tree in East China.No study has evaluated the optimal endogenous reference gene for qRT-PCR analysis in T.miqueliana systematically.In this study,fifteen commonly used reference genes were selected as candidate genes,and the stabilities of their expressions were assessed using four algorithms(GeNorm,NormFiner,BestKeeper,and DeltaCt)in nine experimental datasets.The final integrated evaluation was performed using a comprehensive analysis algorithm(RefFinder).Finally,a target MYB transcription factor gene(TmMYB)was used to verify the accuracy of the candidate reference genes.The results showed that PP2αwas the most stable in tissue set,while RPS13 and SAMCD were optimal for heat and cold stress,respectively.Under waterlogging stress,PP2αand TUB were the most stable genes in the leaves and roots,respectively.EF1αand PP2αwere optimal for drying stress in leaf and root tissues.TUB and ACT7 were the most stable genes in the leaf and root tissues under salt stress.This is the first systematic evaluation of candidate reference genes in T.miqueliana,and it will benefit future studies on expression and functional analysis of target genes in T.miqueliana. | Huanli Wang Lingjun Yan Xi Huang Zhongwei Wang Yuanhao Yue Shijie Tang | 2022 | Phyton-International Journal of Experimental Botany2022，91，10： | 0 |
您所在的机构暂未开通此部分的功能，请联系管理员