|
|
|
题名
|
作者
|
年代
|
出处
|
被引量
|
| 1 | Parallel Learning:a Perspective and a Framework显示文摘The development of machine learning in complex system is hindered by two problems nowadays.The first problem is the inefficiency of exploration in state and action space,which leads to the data-hungry of some state-of-art data-driven algorithm.The second problem is the lack of a general theory which can be used to analyze and implement a complex learning system.In this paper,we proposed a general methods that can address both two issues.We combine the concepts of descriptive learning,predictive learning,and prescriptive learning into a uniform framework,so as to build a parallel system allowing learning system improved by self-boosting.Formulating a new perspective of data,knowledge and action,we provide a new methodology called parallel learning to design machine learning system for real-world problems. | Li Li Yilun Lin Nanning Zheng Fei-Yue Wang | 2017 | IEEE/CAA Journal of Automatica Sinica2017，4，3： | 25 |
| 2 | Generative Adversarial Networks:Introduction and Outlook显示文摘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. | Kunfeng Wang Chao Gou Yanjie Duan Yilun Lin Xinhu Zheng Fei-Yue Wang | 2017 | IEEE/CAA Journal of Automatica Sinica2017，4，4： | 24 |
| 3 | What Does ChatGPT Say:The DAO from Algorithmic Intelligence to Linguistic Intelligence显示文摘THE well-known ancient Chinese philosopher Lao Tzu(老子)or Laozi(6th~4th century BC during the Spring and Autumn period)started his classic Tao Teh Ching《道德经》or Dao De Jing(see Fig.1)with six Chinese characters:'道(Dao)可(Ke)道(Dao)非(Fei)常(Chang)道(Dao)',which has been traditionally interpreted as“道可道,非常道”or'The Dao that can be spoken is not the eternal Dao'. | Fei-Yue Wang Qinghai Miao Xuan Li Xingxia Wang Yilun Lin | 2023 | IEEE/CAA Journal of Automatica Sinica2023，10，3： | 23 |
| 4 | Functional group differentiation of isomeric solvents enables distinct zinc anode chemistry显示文摘Electrolytes hold the key to realizing reliable zinc(Zn)anodes.Divergent organic molecules have been proven effective in stabilizing Zn anodes;however,irrational comparisons exist due to the uncontrolled molecular weights and functional group amounts.In this work,two“isomeric molecules”:1,2-dimethoxyethane(DME)and 1-methoxy-2-propanol(PM),with identical molecular weights but different functional groups,have been studied as co-solvents in electrolytes,which have delivered distinct electrochemical performance.Experimental and simulative study indicates the dipole moment induced by the hydroxyl groups in PM(higher molecular polarity than ether groups in DME)reconstructs the space charge region,enhances the concentration of Zn^(2+)in the vicinity of Zn anodes,and in-situ derives different solid electrolyte interphase(SEI)models and electrode-electrolyte interfaces,resulting in exceptional cycling stability.Remarkably,the Zn||Cu cell with PM worked over 2000 cycles with high Coulombic efficiency(CE)of 99.7%.The Zn||Zn symmetric cell cycled over 2000 h at 1 mA·cm^(−2),and showed excellent stability at an ultrahigh current density of 10 mA·cm^(−2)and capacity of 20 mAh·cm^(−2)over 200 h(depth of discharge,DOD of 70%).The Zn||sodium vanadate pouch cell with a high mass loading of 6.3 mg·cm^(−2)and a high capacity of 24 mAh demonstrates superior cyclability after 570 h.This work can be a good starting point to provide reliable guidance on electrolyte design for practical aqueous Zn batteries. | Chao Liu Qing Li Yilun Lin Zhiquan Wei Yihan Yang Cuiping Han Minshen Zhu Haiyan Zhang Hongfei Li | 2023 | Nano Research Energy2023，2，2： | 1 |
| 5 | Investigating the dynamic memory effect of human drivers via ON-LSTM显示文摘It is a widely accepted view that considering the memory effects of historical information(driving operations)is beneficial for vehicle trajectory prediction models to improve prediction accuracy.However,many commonly used models(e.g.,long short-term memory,LSTM)can only implicitly simulate memory effects,but lack effective mechanisms to capture memory effects from sequence data and estimate their effective time range(ETR).This shortage makes it hard to dynamically configure the most suitable length of used historical information according to the current driving behavior,which harms the good understanding of vehicle motion.To address this problem,we propose a modified trajectory prediction model based on ordered neuron LSTM(ON-LSTM).We demonstrate the feasibility of ETR estimation based on ON-LSTM and propose an ETR estimation method.We estimate the ETR of driving fluctuations and lane change operations on the NGSIM I-80 dataset.The experiment results prove that the proposed method can well capture the memory effects during trajectory prediction.Moreover,the estimated ETR values are in agreement with our intuitions. | Shengzhe DAI Zhiheng LI Li LI Dongpu CAO Xingyuan DAI Yilun LIN | 2020 | Science China(Information Sciences)2020，63，9： | 1 |
| 6 | Image-based traffic signal control via world models显示文摘Traffic signal control is shifting from passive control to proactive control, which enables the controller to direct current traffic flow to reach its expected destinations. To this end, an effective prediction model is needed for signal controllers. What to predict, how to predict, and how to leverage the prediction for control policy optimization are critical problems for proactive traffic signal control. In this paper, we use an image that contains vehicle positions to describe intersection traffic states. Then, inspired by a model-based reinforcement learning method, DreamerV2,we introduce a novel learning-based traffic world model. The traffic world model that describes traffic dynamics in image form is used as an abstract alternative to the traffic environment to generate multi-step planning data for control policy optimization. In the execution phase, the optimized traffic controller directly outputs actions in real time based on abstract representations of traffic states, and the world model can also predict the impact of different control behaviors on future traffic conditions. Experimental results indicate that the traffic world model enables the optimized real-time control policy to outperform common baselines, and the model achieves accurate image-based prediction, showing promising applications in futuristic traffic signal control. | Xingyuan DAI Chen ZHAO Xiao WANG Yisheng LV Yilun LIN Fei-Yue WANG | 2022 | Frontiers of Information Technology & Electronic Engineering2022，23，12： | 1 |
| 7 | Aerosol oxalate and its implication to haze pollution in Shanghai,China显示文摘A total of 238 samples of PM2.5and TSP were analyzed to study the characteristics,sources,and formation pathways of aerosol oxalate in Shanghai in four seasons of 2007.The concentrations of oxalate were0.07–0.41 lg/m3in PM2.5and 0.10–0.48 lg/m3in TSP,respectively.Oxalate displayed a seasonal variation of autumn[summer[winter[spring in both PM2.5and TSP and was dominantly present in PM2.5in all samples.Correlation between oxalate and K?and high ratio of oxalate/K?suggested that biomass burning was a secondary source of aerosol oxalate in Shanghai,in addition to urban VOCs sources(vehicular and industrial emissions),especially in autumn.Secondary formation accounted for the majority of aerosol oxalate in Shanghai,which was supported by the high correlation of oxalate with nssSO42-,K?and NO3-,proceeding from different mechanisms.Relatively high ambient RH together with high cloud cover was found benefiting the secondary formation of aerosol oxalate.The in-cloud process(aqueous-phase oxidation)was proposed to be likely the major formation pathway of aerosol oxalate in Shanghai,which was supported by the high correlation of oxalate with nss-SO42-and K?,dominant residence of oxalate in droplet mode and result of favorable meteorological condition analysis.High correlation of oxalate and NO3-reflected the OH radical involved oxidation chemistry of the two species in the atmosphere and also suggested that gas-particle surface reactions and the evaporation–condensation process were both possible secondary formation pathways of aerosol oxalate in coarser particle mode([1.0 lm).As a major water-soluble organic compound in aerosols,concentration of oxalate showed a distinct negative correlation to the atmospheric visibility,which implied that aerosol organic compounds could play an important role in haze pollution as well as in air quality in Shanghai. | Yilun Jiang Guoshun Zhuang Qiongzhen Wang Tingna Liu Kan Huang Joshua S.Fu Juan Li Yanfen Lin Rong Zhang Congrui Deng | 2014 | Chinese Science Bulletin2014，59，2： | 1 |
| 8 | Coiled Polymer Artificial Muscles Having Dual-Mode Actuation with Large Stress Generation显示文摘Coiled polymer artificial muscles with both large tensile stroke and giant force generation are needed for practical applications in robotics,soft exosuits,and prosthesis.However,most polymer yarn artificial muscles cannot generate a large force or stress.Here,we report an inexpensive Twisted and Coiled Polymer artificial muscle(TCP)that performs both large isobaric and isometric contractions.This TCP can generate a tensile stroke of 20.1%and a specific work capacity of up to 1.3 kJ kg^(−1) during temperature changes from 20 to 180℃.Moreover,the nylon yarn artificial muscle produced a reversible output stress of 28.4 MPa,which is 100 times larger than human skeletal muscle.A robot arm and a simple gripper were made to demonstrate the isobaric actuation and isometric actuation of our TCP muscle,repectivley.Thus,the polymer artificial muscles with dual-mode actuation show potential applications in the field of robotics,grippers,and exoskeletons and so on. | Xinghao Hu Runmin Liu Kai Zhao Yilun Wang Xianfu Bao Lin Xu Guanggui Cheng Jianning Ding | 2023 | Journal of Bionic Engineering2023，20，4： | 0 |
您所在的机构暂未开通此部分的功能，请联系管理员