Prof. Gary G. Yen
Oklahoma State University, U.S.A.
Gary G. Yen received the Ph.D. degree in electrical and computer engineering from the University of Notre Dame in 1992. He is currently a Regents Professor in the School of Electrical and Computer Engineering, Oklahoma State University. His research interest includes intelligent control, computational intelligence, evolutionary multiobjective optimization, conditional health monitoring, signal processing and their industrial/defense applications.
Gary was an associate editor of the IEEE Transactions on Neural Networks and IEEE Control Systems Magazine during 1994-1999, and of the IEEE Transactions on Control Systems Technology, IEEE Transactions on Systems, Man and Cybernetics and IFAC Journal on Automatica and Mechatronics during 2000-2010. He is currently serving as an associate editor for the IEEE Transactions on Evolutionary Computation and IEEE Transactions on Cybernetics. Gary served as Vice President for the Technical Activities, IEEE Computational Intelligence Society in 2004-2005 and is the founding editor-in-chief of the IEEE Computational Intelligence Magazine, 2006-2009. He was the President of the IEEE Computational Intelligence Society in 2010-2011 and is elected as a Distinguished Lecturer for the term 2012-2014. He received Regents Distinguished Research Award from OSU in 2009, 2011 Andrew P Sage Best Transactions Paper award from IEEE Systems, Man and Cybernetics Society, 2013 Meritorious Service award from IEEE Computational Intelligence Society and 2014 Lockheed Martin Aeronautics Excellence Teaching award. Currently he serves as the chair of IEEE/CIS Fellow Committee and General Co-Chair of 2016 IEEE World Congress on Computational Intelligence to be held in Vancouver, Canada. He is a Fellow of IEEE and IET.
Title: State-of-the-Art Many-Objective Evolutionary Algorithms for Optimization
Gary G. Yen, Regents Professor, FIEEE, FIET
Oklahoma State University
Abstract: Evolutionary computation is the study of biologically motivated computational paradigms which exert novel ideas and inspiration from natural evolution and adaptation. The applications of population-based heuristics in solving multiobjective optimization problems have been receiving a growing attention. To search for a family of Pareto optimal solutions based on nature-inspiring problem solving paradigms, Evolutionary Multiobjective Optimization Algorithms have been successfully exploited to solve optimization problems in which the fitness measures and even constraints are uncertain and changed over time.
When encounter optimization problems with many objectives, nearly all designs performs poorly because of loss of selection pressure in fitness evaluation solely based upon Pareto optimality principle. This talk will survey recently published literature along this line of research- evolutionary algorithm for many-objective optimization and its real-world applications. In particular, focus will be placed on the design of selection strategy, including mating selection and environmental selection. We will show the design of a coordinated selection strategy to improve the performance of evolutionary algorithms in many-objective optimization. This selection strategy considers three crucial factors: 1) the new mating selection criterion considers both the quality of each selected parent and the effectiveness of the combination of selected parents; 2) the new environmental selection criterion directly focuses on the performance of the whole population rather than single individual alone, and 3) both selection strategies are complement to each other and the coordination between them in the evolutionary process can achieve a better performance than each of them used individually. Based on performance metrics ensemble, we will provide a comprehensive measure among all competitors and more importantly reveal insight pertaining to specific problem characteristics that the underlying evolutionary algorithm could perform the best.
Kay Chen Tan (SM’05) (SM’08–F’14) received the B.Eng. (Hons.) degree in electronics and electrical engineering and the Ph.D. degree from University of Glasgow, Glasgow, U.K., in 1994 and 1997, respectively. He is an Associate Professor with the Department of Electrical and Computer Engineering, National University of Singapore, Singapore. He is actively pursuing research in computational and artificial intelligence, with applications to multi-objective optimization, scheduling, data analytics, prognostics, BCI etc.
Dr Tan has published over 120 journal papers, over 120 papers in conference proceedings, co-authored 5 books. He has been an Invited Keynote/Plenary speaker for over 50 international conferences. He was the General Co-Chair for IEEE Congress on Evolutionary Computation 2007 in Singapore and is the General Co-Chair for IEEE World Congress on Computational Intelligence 2016 in Vancouver, Canada. Dr Tan is currently an elected member of AdCom (2014-2016) and is an IEEE Distinguished Lecturer of IEEE Computational Intelligence Society (2011-2013; 2015-2017).
Dr Tan is a Fellow of IEEE. He is also the Editor-in-Chief of IEEE Transactions on Evolutionary Computation. He served as the Editor-in-Chief of IEEE Computational Intelligence Magazine (2010-2013), and currently serves as an Associate Editor / Editorial Board member of over 20 international journals, such as IEEE Transactions on Cybernetics, IEEE Transactions on Computational Intelligence and AI in Games, Evolutionary Computation (MIT Press), European Journal of Operational Research, Neural Computing and Applications, Journal of Scheduling, International Journal of Systems Science, etc.
He is the awardee of the 2012 IEEE Computational Intelligence Society (CIS) Outstanding Early Career Award for his contributions to evolutionary computation in multi-objective optimization. He also received the 2016 IEEE CIS Outstanding TNNLS Paper Award for his paper titled "Rapid Feedforward Computation by Temporal Encoding and Learning with Spiking Neurons". He also received the Recognition Award (2008) from the International Network for Engineering Education & Research (iNEER) for his outstanding contributions to engineering education and research. He was felicitated by the International Neural Network Society (INNS) India Regional Chapter (2014) for his outstanding contributions in the field of computational intelligence.
Title: On Prognostics using Evolutionary Multi-objective Optimization
Abstract: Multi-objective optimization is widely found in many fields, such as logistics, economics, engineering, or whenever optimal decisions need to be made in the presence of trade-offs. The problem is challenging because it involves the simultaneous optimization of several conflicting objectives in the Pareto optimal sense and requires researchers to address many issues that are unique to MO problems. This talk will provide an overview of evolutionary computation for multi-objective optimization (EMO). It will then present various applications of EMO for solving engineering problems particularly in the area of robust prognostic. As one of the key enablers of condition based maintenance, prognostic involves the core task of determining the remaining useful life (RUL) of the system. This talk will discuss the use of neural network ensembles to improve the prediction accuracy of RUL estimation as well as the use of EMO to optimize the ensemble hyper-parameters. A case study involving the estimation of RUL for turbofan engines will also be presented in the talk.
Tshilidzi Marwala (OMB) born 28 July 1971 in Venda, Transvaal, South Africa is the currently the Deputy Vice Chancellor: Research, Innovation, Postgraduate Studies and the Library at the University of Johannesburg. Marwala was previously a Dean of Engineering at the University of Johannesburg, a Professor of Electrical Engineering, the Carl and Emily Fuchs Chair of Systems and Control Engineering as well as the DST/NRF South Africa Research Chair of Systems Engineering at the University of the Witwatersrand. He holds a Bachelor of Science in Mechanical Engineering (Magna Cum Laude) from Case Western Reserve University in USA, a Master of Engineering from the University of Pretoria, a PhD in Engineering from the University of Cambridge. He was a post-doctoral research associate at the Imperial College of Science, Technology and Medicine and in year 2006 to 2007 was a visiting fellow at Harvard University. In the year 2007 to 2008, he has been appointed a visiting fellow of Wolfson College, Cambridge. He has supervised 47 masters and 19 PhD students to completion and has published over 300 papers and 8 books. He is a fellow of TWAS, The World Academy of Sciences, Academy of Science of South Africa and African Academy of Sciences as well as a senior member of the IEEE and a distinguished member of the Association for Computing Machinery. His work has appeared in publications such as the New Scientist.
Title: Evolutionary computation and deep learning for missing data estimation in large databases (by Tshilidzi Marwala, Collins Leke and Evan Hurwitz)
Abstract: Recent advances in computational and artificial intelligence techniques have rekindled interest in the domain of missing data imputation. The techniques in the domain include amongst others: Expectation Maximization, Neural Networks with Evolutionary Algorithms or optimization techniques, K-Nearest Neighbor approaches and most recently, Extreme Machine Learning with Gaussian Mixture Models. The presence of missing data entries in databases render the tasks of decision-making and data analysis as well as prediction and classification tasks nontrivial. As a result this area has attracted a lot of research interest with the aim being to yield accurate and time efficient missing data imputation techniques especially when time sensitive applications are concerned like power plants and winding processes. The high-dimensional nature of most real life datasets nowadays makes imputing the missing data entries more difficult using the existing techniques. In this article, considering arbitrary and monotone missing data patterns, we introduce novel missing data imputation methods that implement deep neural networks built using autoencoders and denoising autoencoders in conjunction with evolutionary computation techniques such as the social spider and fireworks optimization algorithms on high-dimensional data with varying degrees of missing-ness. Also considered are the missing at random, missing completely at random and missing not at random missing data mechanisms. The deep neural networks are initialized with pre-trained weights from Restricted Boltzmann Machines prior to back-propagation being implemented to update the network weights and biases. The idea behind this is the weights are initialized in a good solution space compared to that which are obtained should the weights be initialized randomly. These novel techniques are compared against shallow and deep multilayer perceptron neural networks with randomly initialized weights to observe whether there is an advantage to using pre-trained networks in a deep architecture setting on the missing data imputation task.
To be added.