检索结果-南通市图书馆

Feature Selection and Feature Learning for High-dimensional Batch Reinforcement Learning: A Survey

在线全文

学校读者我要写书评

暂无评论

International Journal of Automation and computing 2015年第3期12卷 229-242页

作者： De-Rong Liu Hong-Liang Li Ding Wang State Key laboratory of management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences

Tremendous amount of data are being generated and saved in many complex engineering and social systems every *** is significant and feasible to utilize the big data to make better decisions by machine learning techniques. In this paper, we focus on batch reinforcement learning(RL) algorithms for discounted Markov decision processes(MDPs) with large discrete or continuous state spaces, aiming to learn the best possible policy given a fixed amount of training data. The batch RL algorithms with handcrafted feature representations work well for low-dimensional MDPs. However, for many real-world RL tasks which often involve high-dimensional state spaces, it is difficult and even infeasible to use feature engineering methods to design features for value function approximation. To cope with high-dimensional RL problems, the desire to obtain data-driven features has led to a lot of works in incorporating feature selection and feature learning into traditional batch RL algorithms. In this paper, we provide a comprehensive survey on automatic feature selection and unsupervised feature learning for high-dimensional batch RL. Moreover, we present recent theoretical developments on applying statistical learning to establish finite-sample error bounds for batch RL algorithms based on weighted Lpnorms. Finally, we derive some future directions in the research of RL algorithms, theories and applications.

关键词： Intelligent control reinforcement learning adaptive dynamic programming feature selection feature learning big data.

Crowd Energy management: New Paradigm for Electricity Market

在线全文

学校读者我要写书评

暂无评论

Journal of Electronic Science and Technology 2015年第3期13卷 195-205页

作者： Mario Gstrein Stephanie Teufel the International Institute of management in Technology University of Fribourg

The traditionally centralized approach of electricity networks is progressively undergoing a shift towards a decentralized, distributed structure. The local and crowd-based principle is transforming the existing supply chain and related activities into a value network （VN）. Previous researches on crowd value network concepts focus on the activities of infrastructure and load management and neglect activities that generate collaboration. Collaboration with and within crowds particularly demands a different mindset and management of sharing values, information, benefit, and risks. Furthermore, these concepts must integrate technical, processual, and social aspects. Thus, this article proposes a holistic framework of electricity VN management for crowd energy. It redefines VN activities for infrastructure and load management while appending VN activities for social electricity handling. Additionally, the framework illustrates the interactions among these three elements and concludes with an adaption cycle for the crowd value network.

关键词： Crowd distribution systems value electricity management. energy decentralized network collaboration,

Performance Evaluation of Machine Learning Methods in Cultural Modeling

在线全文

学校读者我要写书评

暂无评论

Journal of Computer Science & Technology 2009年第6期24卷 1010-1017页

作者：李晓晨毛文吉曾大军苏鹏王飞跃 Key laboratory of Complex Systems and Intelligence Science Institute of Automation Chinese Academy of Sciences Department of management Information Systems University of Arizona School of management Engineering Shandong Jianzhu University

Cultural modeling （CM） is an emergent and promising research area in social computing. It aims to develop behavioral models of human groups and analyze the impact of culture factors on human group behavior using computational methods. Machine learning methods, in particular classification, play a critical role in such applications. Since various cultural-related data sets possess different characteristics, it is important to gain a computational understanding of performance characteristics of various machine learning methods. In this paper, we investigate the performance of seven representative classification algorithms using a benchmark cultural modeling data set and analyze the experimental results as to group behavior forecasting.

关键词： cultural modeling (CM) machine learning classification group behavior forecasting

Photodynamic therapy guidelines for the management of oral leucoplakia

在线全文

学校读者我要写书评

暂无评论

International Journal of Oral Science 2019年第2期11卷 103-107页

作者： Qianming Chen Hongxia Dan Fan Tang Jiongke Wang Xiaoying Li Junxin Cheng Hang Zhao Xin Zeng State Key laboratory of Oral Diseases & National Clinical Research Center for Oral Diseases & Chinese Academy of Medical Sciences Research Unit of Oral Carcinogenesis and management West China Hospital of Stomatology Sichuan University

With recent developments in photosensitizers and light delivery systems,topical 5-aminolevulinic acid-mediated photodynamic therapy (ALA-PDT) has become the fourth alternative therapeutic approach in the management of oral leucoplakia (OLK) due to its minimally invasive nature,efficacy,and low risk of systemic side effects and *** report presents step-by-step guidelines for applying topical ALA-PDT in the management of OLK based on both the clinical experience of the authors and a systematic review of the current *** using protocols with standardized parameters and randomized clinical trials at multiple centres with adequate sample sizes and both interim and long-term follow-ups are needed before universally applicable guidelines can be produced in this field.

关键词： management oral leucoplakia photosensitizers ALA-PDT

Constructing an assessment index system for strategic risk management in coal science and technology enterprises

在线全文

学校读者我要写书评

暂无评论

International Journal of Mining Science and Technology 2016年第4期26卷 653-660页

作者： Fan Baoying Yuan Yingju School of management China University of Mining and Technology

Coal occupies a dominant position in China＇s national economy and is an essential energy source for future industrial development. To inform the efficient and sustainable development of the coal industry, this paper analyzes the sources of strategic risk for coal science and technology enterprises using 3 first-class indicators, 10 second-level indicators, and 37 observation points established through the existing research literature and experience. Moreover, in accordance to the obtained initial index data, the selected indicators have been tested and screened using reliability and membership degree analyses to remove redundant variables, avoid juxtaposition of risk factors at different levels, and reduce the influ- ence of some tiny risk factors for enterprise strategic risk. Then, factor analysis of external environment factor sub-scale was carried out. Factors are extracted according to a standard characteristic value greater than 1. Variables with high coefficients are classified into one factor category; and finally, 3 first-class indicators. 8 second-level indicators, and 37 observation noints are reconstructed.

关键词： Coal science and technology enterprises Strategic risk Risk assessment index Constructing

Microsatellites reveal strong genetic structure in the common cutworm, Spodoptera litura

在线全文

学校读者我要写书评

暂无评论

Journal of Integrative Agriculture 2019年第3期18卷 636-643页

作者： WU Huai-heng WAN Peng HUANG Min-song LEI Chao-liang Hubei Insect Resources Utilization and Sustainable Pest management Key laboratory Huazhong Agricultural University Key laboratory of Integrated Pest management on Crops in Central China Institute of Plant Protection and Soil ScienceHubei Academy of Agricultural Sciences

The common cutworm, Spodoptera litura(Lepidoptera: Noctuidae), is a voracious agricultural pest. To increase understanding of the migration patterns and genetic diversity of different geographic populations of this species, we analyzed genetic variation in nine microsatellite loci among 576 individuals collected from 17 locations in China and one in Myanmar during 2011–2012. We successfully identified 162 alleles, with an average of 18 alleles per locus and a range of 5 to 34. The mean observed heterozygosity of the 18 populations ranged from 0.18 to 0.98, and the expected heterozygosity ranged from 0.19 to 0.89. For the nine microsatellite markers studied, polymorphism information content ranged from 0.18 to 0.88(mean=0.64). We found low genetic differentiation among the 18 populations(mean F-statistics(F_(ST))=0.05) and high genetic diversity among individuals. Principle coordinates analysis indicated no significant correlation between geographic and genetic distance(r=0.04). The value of N_m(N_m>4) estimated using coalescent-based methods suggests strong gene flow with migration. The nine microsatellite markers identified in this study will be beneficial for further investigation of migration patterns and genetic diversity in S. litura.

关键词： Spodoptera litura microsatellite genetic diversity

Eicosanoids mediate nodulation reactions to bacterial Escherichia coil K 12 infections in larvae of the oriental blowfly, Chrysomya megacephala

在线全文

学校读者我要写书评

暂无评论

Insect Science 2009年第5期16卷 387-392页

作者： Fu Zhao Bing Chen Yong Wang Fen Zhu Chao-Liang Lei Hubei Insect Resources Utilization and Sustainable Pest management Key laboratory Huazhong Agricultual University Wuhan China State Key laboratory of Integrated management of Pest Insects and Rodents Institute of Zoology Chinese Academy of Sciences BeijingChina

Nodulation is the predominant cellular defense reaction to bacterial challenges in insects. In this study, third instar larvae of Chrysomya megacephala were injected with bacteria, Escherichia coli K 12 （10^6 CFU/mL, 2μL）, immediately prior to injection of inhibitors of eicosanoid biosynthesis, which sharply reduced nodulation response. Test larvae were treated with specific inhibitors ofphospholipase A2 （dexamethasone）, cyclo- oxygenase （indomethacin, ibuprofen and piroxicam）, dual cyclo-oxygenase/lipoxygenase （phenidone） and lipoxygenase （esculetin） and these reduced nodulation except esculetin. The influence of bacteria was obvious within 2 h of injection （5 nodules/larva）, and increased to a maximum after 8 h （with 15 nodules/larva）, and then significantly reduced over 24 h （9 nodules/larva）. The inhibitory influence of dexamethasone was apparent within 2 h of injection （4 vs. 5 nodules/larva）, and nodulation was significantly reduced, compared to control, over 24 h （5 vs. 8 nodules/larva）. Increased dosages of ibuprofen, indomethacin, piroxicam and phenidone led to decreased numbers of nodules. Nodules continued to exist during the pupal stage. However, the effects of dexamethasone were reversed by treating bacteria-injected insects with an eicosanoid-precursor polyunsaturated fatty acid, arachidonic acid. These findings approved our view that eicosanoid can mediate cellular defense mechanisms in response to bacterial infections in another Dipteran insect C. rnegacephala.

关键词： bacteria Chrysomya megacephala eicosanoids Escherichia coli K12 nodulation

Boosting cationic and anionic redox activity of Li-rich layered oxide cathodes via Li/Ni disordered regulation

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Energy Chemistry 2025年第1期 533-543页

作者： Zewen Liu Zhen Wu Hao Wang Xudong Zhang Yuanzhen Chen Yongning Liu Shengwu Guo Shenghua Chen Yanli Nan Yan Liu State Key laboratory for Mechanical Behavior of Materials Xi'an Jiaotong University School of Materials Science and Engineering Xi'an University of Architecture and Technology Hefei Advanced Computing Center Operation management Corp Ltd Network Information Center Xi'an Jiaotong University

Lithium-rich layered oxides （LLOs） are increasingly recognized as promising cathode materials for nextgeneration high-energy-density lithium-ion batteries （LIBs）.However,they suffer from voltage decay and low initial Coulombic efficiency （ICE） due to severe structural degradation caused by irreversible O ***,we introduce a three-in-one strategy of increasing Ni and Mn content,along with Li/Ni disordering and TM–O covalency regulation to boost cationic and anionic redox activity simultaneously and thus enhance the electrochemical activity of *** target material,Li1.2Ni0.168Mn0.558Co0.074O2（L1）,exhibits an improved ICE of 87.2%and specific capacity of 293.2 mA h g-1and minimal voltage decay of less than 0.53 m V cycle-1over 300 cycles at 1C,compared to Li1.2Ni0.13Mn0.54Co0.13O2（Ls）(274.4 mA h g-1for initial capacity,73.8%for ICE and voltage decay of 0.84 mV/cycle over 300 cycles at 1C).Theoretical calculations reveal that the density of states （DOS） area near the Fermi energy level for L1 is larger than that of Ls,indicating higher anionic and cationic redox reactivity than ***,L1 exhibits increased O-vacancy formation energy due to higher Li/Ni disordering of 4.76%（quantified by X-ray diffraction Rietveld refinement） and enhanced TM–O covalency,making lattice O release more difficult and thus improving electrochemical *** increased Li/Ni disordering also leads to more Ni2+presence in the Li layer,which acts as a pillar during Li+de-embedding,improving structural *** research not only presents a viable approach to designing low-Co LLOs with enhanced capacity and ICE but also contributes significantly to the fundamental understanding of structural regulation in high-performance LIB cathodes.

关键词：

Real-time anomaly detection for very short-term load forecasting

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Modern Power Systems and Clean Energy 2018年第2期6卷 235-243页

作者： Jian LUO Tao HONG Meng YUE School of management Science and Engineering Dongbei University of Finance and Economics Department of Systems Engineering and Engineering management University of North Carolina at Charlotte Brookhaven National laboratory

Although the recent load information is critical to very short-term load forecasting(VSTLF), power companies often have difficulties in collecting the most recent load values accurately and timely for VSTLF *** paper tackles the problem of real-time anomaly detection in most recent load information used by *** paper proposes a model-based anomaly detection method that consists of two components, a dynamic regression model and an adaptive anomaly threshold. The case study is developed using the data from ISO New England. This paper demonstrates that the proposed method significantly outperforms three other anomaly detection methods including two methods commonly used in the field and one state-of-the-art method used by a winning team of the Global Energy Forecasting Competition 2014. Finally, a general anomaly detection framework is proposed for the future research.

关键词： Real-time anomaly detection Very short-term load forecasting Multiple linear regression Data cleansing