Title：Data-driven evolutionary optimization：A taxonomy and case studies
Professor Yaochu Jin, University of Surrey, United Kingdom.
Biography: Yaochu Jin received the B.Sc., M.Sc., and Ph.D. degrees from Zhejiang University, Hangzhou, China, in 1988, 1991, and 1996, respectively, and the Dr.-Ing. degree from Ruhr University Bochum, Germany, in 2001.
He is a Professor in Computational Intelligence, Department of Computer Science, University of Surrey, Guildford, U.K., where he heads the Nature Inspired Computing and Engineering Group. He is also a Finland Distinguished Professor funded by the Finnish Agency for Innovation (Tekes) and a Changjiang Distinguished Visiting Professor appointed by the Ministry of Education, China. His main research interests include data-driven surrogate-assisted evolutionary optimization, evolutionary multi-objective optimization, evolutionary learning, interpretable and secure machine learning, and evolutionary developmental systems. He has (co)authored over 250 peer-reviewed journal and conference papers and been granted eight patents on evolutionary optimization. He has delivered 30 invited keynote speeches at international conferences.
Dr Jin is the Editor-in-Chief of the IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS and Co-Editor-in-Chief of Complex & Intelligent Systems. He is an IEEE Distinguished Lecturer (2013-2015 and 2017-2019) and past Vice President for Technical Activities of the IEEE Computational Intelligence Society (2014-2015). He is the recipient of the 2018 IEEE Transactions on Evolutionary Computation Outstanding Paper Award, the 2015 and 2017 IEEE Computational Intelligence Magazine Outstanding Paper Award, and the Best Paper Award of the 2010 IEEE Symposium on Computational Intelligence in Bioinformatics and Computational Biology. He is a Fellow of IEEE.
Personal webpage: https://www.surrey.ac.uk/people/yaochu-jin
Title：Modeling and evolutionary heuristic solutions for some operations planning problems
Professor Jiyin Liu, Loughborough University, United Kingdom
Biography: Jiyin Liu is a Professor of Operations Management in the School of Business and Economics at Loughborough University. He works in both operations management and operational research areas. Jiyin received his PhD in Manufacturing Engineering and Operations Management from the University of Nottingham in 1993 and lectured at Hong Kong University of Science and Technology before joining Loughborough at the end of 2003. He is also a Chang Jiang Scholar Chair Professor at Northeastern University of China since 2007.
Jiyin’s research is mainly on modelling and optimisation of operations planning problems in logistics and production systems. His research outputs have been published in leading academic journals in the areas of operational research and operations management, such as Operations Research, European Journal of Operational Research, Transportation Research Part B, Manufacturing & Service Operations Management, Naval Research Logistics, International Journal of Production Research, IIE Transactions, and IEEE Transactions.
Jiyin focuses on problems with both academic significance and practical relevance. His work has been supported by research funding agencies and by industry. He has collaborated with companies such as Baosteel, British Telecom, Hongkong International Terminals, Hong Kong Air Cargo Terminals Limited, and Philips Electronics. He received the Franz Edelman Finalist Awards from INFORMS for Achievements in Practice of Operations Research and Management Sciences twice for works on decision support in container terminal operations (2004) and in steel industry (2013), respectively.
Personal webpage: http://www.lboro.ac.uk/departments/sbe/staff/jiyin-liu/
Title：Implementing Neuromorphic Reservoir Computing with Self-Assembled Memristive Switching Networks
Professor Thomas H. LaBean, North Carolina State University, USA
Abstract: Training and use of large nonlinear neural networks on conventional computer architectures is impeded by poor scalability and high energy penalties for sequential updates of neuron weights. Here we develop a low power and highly scalable computing paradigm of self-assembled neural network-like architectures using DNA origami, functional peptides, and inorganic components for fabrication of circuits with potential for real-time computing. We experimentally and theoretically examine circuits created by the molecular assembly of functional components capable of displaying complex, emergent electronic behaviors such as memristor based reservoir computing. Deterministic assembly at low nanometer length-scales followed by stochastic assembly at high nanoscale and up to micron scale should provide circuits with exploitable electronic behaviors. Theoretical work focuses particularly on modeling and simulation of device function and network structure/function in order to predict emergent electronic properties. Network architectures will follow neuromorphic principles and will result in trainable or learnable circuits with potential capabilities including memory, logic, and complex signal processing.
Biography: Thomas H. LaBean is Professor of Materials Science and Engineering at North Carolina State University. He earned BS and PhD degrees in Biochemistry from the Honors College at Michigan State University and the University of Pennsylvania, respectively. He studied folding and assembly of arbitrary sequence proteins in graduate school, then moved to Duke University as a Biochemistry postdoc and studied de novo protein design. As a Research Professor in Computer Science, he worked on DNA-based molecular computation and self-assembling biomolecular nanostructures. He has been at North Carolina State University since 2011, and his current research involves self-assembling polypeptides and DNA nanostructures for molecular materials, bioinspired nanoelectronics fabrication, and nanomedicine.
Personal webpage: https://www.mse.ncsu.edu/profile/thlabean
Title：Production, Logistics and Energy Optimization and Application in Steel Industry
Professor Lixin Tang，Northeastern University，China
Abstract: This talk discusses some interesting topics on the scheduling and data analytics of production, logistics and energy in the steel industry, including: 1) production scheduling in steel-making and hot/cold rolling operations; 2) logistics scheduling in storage/stowage, shuffling, transportation and (un)loading operations; 3) energy optimization including energy allocation and coordinated planning and scheduling of production and energy; 4) data based analytics including dynamic analytics of BOF steelmaking process based on multi-stage modeling; temperature prediction of blast furnace; temperature prediction of molten iron in transportation process; energy analytics for estimation, prediction of generation and consumption, diagnosis and benchmarking; temperature prediction of reheat furnace based on mechanism and data; strip quality analytics of continuous annealing based on multi-objective ensemble learning; process monitoring and diagnosis of continuous annealing based on mechanism and data.
Biography: Lixin Tang is a Cheung Kong Scholars Chair Professor, the Vice President of Northeastern University, the Director of the Institute of Industrial & Systems Engineering, and the Head of the Operation Analytics and Optimization Centre for Smart Industry at Northeastern University of China.
His research interests cover plant-wide production and logistics planning, production and logistics batching and scheduling, operations analytics and optimization for smart industry, convex and integer optimization, data analytics and machine learning, computational intelligent optimization and engineering applications in manufacturing (steel, petroleum-chemical, nonferrous), energy, resources industry and logistics systems.
He has published 106 papers in international journals such as OR, M&SOM, INFORMS Journal on Computing, IISE Transactions, NRL, IEEE Transactions on Evolutionary Computation. He was selected into the list of 2014, 2015 and 2016 Most Cited Chinese Researchers by Elsevier. The paper published on flagship journal IIE Transactions (now renamed as IISE Transactions) won the Best Applications Paper Award of 2015-2016.
He serves as an Associate Editor of IISE Transactions, IEEE Transactions on Evolutionary Computation, IEEE Transactions on Cybernetics, IEEE Transactions on Automation Science and Engineering, Journal of Scheduling, International Journal of Production Research, Journal of the Operational Research Society, in Editorial Board of Annals of Operations Research, and an Area Editor of the Asia-Pacific Journal of Operational Research.
Title：Decomposition Based Multiobjective Evolutionary Computation: the current state and future
Professor Qingfu Zhang ，City University of Hong Kong，China
Biography: Professor, IEEE Fellow, Department of Computer Science, City University of Hong Kong, Hong Kong
Changjiang Visiting Chair Professor at Xidian University, China, 2011.
Expert in one thousand talent program of China, 2015.
Highly cited researcher in computer science, 2016, 2017
Personal homepage: http://www.cs.cityu.edu.hk/~qzhang/index.html
Title：Generalization and overfitting in deep reinforcement learning
Professor Julian Togelius, New York University, USA
Abstract: Reinforcement learning is the study of methods for learning to act based on interactions with the environment alone. It carries the promise of learning to solve hard problems we do not currently know how to solve, helping to realize the dream of self-learning AI. The last few years have seen reinforcement learning algorithms combined with deep neural networks learn to play an array of games of varying character and complexity. But we’ve also seen some of the limitations of these algorithms. They seem to learn brittle policies, that only work for the particular games, and sometimes even only for particular levels of those games. Why is this happening, and what could we do to about it? I will discuss some of the problems with reinforcement learning research, and showcase work from my lab on characterizing and ameliorating the ills of reinforcement learning. This includes methods for teaching networks gradually more general skills, and for training extremely small networks capable of playing complex games. A central insight is that the environment and training regime is at least as important as the learning algorithm.
Biography: Julian Togelius is an Associate Professor in the Department of Computer Science and Engineering, New York University, USA. He works on artificial intelligence for games and games for artificial intelligence. His current main research directions involve search-based procedural content generation in games, general video game playing, player modeling, generating games based on open data, and fair and relevant benchmarking of AI through game-based competitions. He is the Editor-in-Chief of IEEE Transactions on Games, and has been chair or program chair of several of the main conferences on AI and Games. Togelius holds a BA from Lund University, an MSc from the University of Sussex, and a PhD from the University of Essex. He has previously worked at IDSIA in Lugano and at the IT University of Copenhagen.
Title：Tackling Many Objectives
Abstract: Many optimisation problems in the real world need to consider multiple conflicting objectives simultaneously. Evolutionary algorithms are excellent candidates for finding a good approximation to the Pareto optimal front in a single run. However, many multi-objective optimisation algorithms are effective for two or three objective only. It is an on-going challenge to deal with a larger number of objectives. In this talk, I will explain several methods for dealing with many objectives. First, we will describe a method for reducing a large number of objectives to a smaller one, especially when there is redundancy among different objectives. Second, alternative dominance relationship, other than the Pareto dominance, will be introduced into to make previously non-comparable solutions comparable. Lastly, new algorithms will be introduced to cope with many objectives through the use of two separate archives, for convergence and diversity, respectively. Our studies show that these methods are very effective and outperform other popular methods in the literature.
Biography: Xin Yao is a Chair Professor of Computer Science at the Southern University of Science and Technology, Shenzhen, China, and a part-time Chair Professor of Computer Science at the University of Birmingham, UK. He is an IEEE Fellow, a former President (2014-15) of IEEE Computational Intelligence Society, and a former Editor-in-Chief (2003-08) of IEEE Transactions on Evolutionary Computation. His major research interests include evolutionary computation, ensemble learning and search-based software engineering. His work won the 2001 IEEE Donald G. Fink Prize Paper Award, 2010, 2016 and 2017 IEEE Transactions on Evolutionary Computation Outstanding Paper Awards, 2010 BT Gordon Radley Award for Best Author of Innovation (Finalist), 2011 IEEE Transactions on Neural Networks Outstanding Paper Award, and many other best paper awards. He received the prestigious Royal Society Wolfson Research Merit Award in 2012 and the IEEE CIS Evolutionary Computation Pioneer Award in 2013.