Prof. Lazim Abdullah School of Informatics and Applied Mathematics, Universiti Malaysia Terengganu, Malaysia
Biography: Lazim Abdullah is a professor of computational mathematics at the School of Informatics and Applied Mathematics, Universiti Malaysia Terengganu. He holds a B.Sc (Hons) in Mathematics from the University of Malaya, Kuala Lumpur in June 1984 and the M.Ed in Mathematics Education from University Sains Malaysia, Penang in 1999. He received his Ph.D. from the Universiti Malaysia Terengganu, (Information Technology Development) in 2004. His research focuses on the mathematical theory of fuzzy sets and its applications to social ecology, environmental sciences, health sciences, and manufacturing engineering. His research findings have been published in over two hundred and fifty publications, including refereed journals, conference proceedings, chapters in books, and research books. Currently, he is Director, Academic Planning, Development and Quality of his University and a member of editorial boards of several international journals related to computing and applied mathematics. He is also a regular reviewer for a number of local and international impact factor journals, member of scientific committees of several symposia and conferences at national and international levels. Dr Abdullah is an associate member, IEEE Computational Intelligence Society, a member of the Malaysian Mathematical Society and a member of the International Society on Multiple Criteria Decision Making.
Speech Title: A New Integrated SAW-TOPSIS based on Interval Type-2 Fuzzy Sets for Decision Making Abstract: Uncertainty and fuzziness of the real world problem could be represented by interval type-2 fuzzy sets where additional degrees of flexibility in decision making environment are presumed. This paper aims to propose an integrated method of interval type-2 fuzzy simple additive weighting (IT2 FSAW) and interval type-2 fuzzy technique for order preference by similarly to ideal solution (IT2 FTOPSIS). The IT2 FSAW is used to determine weight for each criterion, while IT2 FTOPSIS method is used to obtain the final ranking for the attributes. A numerical example is used to illustrate the proposed method. In essence, the integrated method is equipped with interval type-2 fuzzy sets in contrast to type-1 fuzzy sets.

	Dr. Wenwu Wang Centre for Vision Speech and Signal Processing, Department of Electronic Engineering, University of Surrey, UK
Biography: Wenwu Wang (M'02–SM'11) received the B.Sc. degree in automatic control, M.E. degree in control science and control engineering, and Ph.D. degree in navigation guidance and control from Harbin Engineering University, Harbin, China, in 1997, 2000, and 2002, respectively. He joined Kings College, London, U.K., in May 2002, as a Postdoctoral Research Associate and transferred to Cardiff University, Cardiff, U.K., in January 2004. In May 2005, he joined the Tao Group Ltd. (now Antix Labs Ltd.), Reading, U.K., as a DSP Engineer. In September 2006, he joined Creative Labs Ltd., Egham, U.K., as an R&D Engineer. Since May 2007, he has been with the Centre for Vision Speech and Signal Processing, University of Surrey, Guildford, U.K., where he is currently a Reader in Signal Processing and a Co-Director of the Machine Audition Laboratory. In 2008, he was a Visiting Scholar with the Perception and Neurodynamics Laboratory and the Center for Cognitive Science, The Ohio State University, Columbus, OH, USA. He has authored or coauthored over 150 publications in these areas, including Machine Audition: Principles, Algorithms and Systems (IGI Global, 2010) and Blind Source Separation: Advances in Theory, Algorithms and Applications (Springer, 2014). His current research interests include blind signal processing, sparse signal processing, audio-visual signal processing, machine learning and perception, machine audition (listening), and statistical anomaly detection. Dr. Wang is currently an Associate Editor for the IEEE TRANSACTIONS ON SIGNAL PROCESSING. He is a Member of the Ministry of Defence University Defence Research Collaboration in Signal Processing (since 2009), a Member of the BBC Audio Research Partnership (since 2011), and an Associate Member of the Surrey Centre for Cyber Security (since 2014). He is (or has been) a Chair, Session Chair, or Technical/Program Committee Member for a number of international conferences, including Publication Co-Chair of ICASSP 2019 (to be held in Brighton, UK), Session Chair of ISP 2015, Local Arrangement Co-Chair of MLSP 2013, Session Chair of ICASSP 2012, Area and Session Chair of EUSIPCO 2012, and Track Chair and Publicity Co-Chair of SSP 2009. He was a Tutorial Speaker for ICASSP 2013 and UDRC Summer School 2014, 2015 and 2016, and SpaRTan/MacSeNet Spring School 2016.
Speech Title: Sparse Analysis Model Based Dictionary Learning and Signal Reconstruction Abstract: Sparse representation has been studied extensively in the past decade in a variety of applications, such as denoising, source separation and classification. Earlier effort has been focused on the well-known synthesis model, where a signal is decomposed as a linear combination of a few atoms of a dictionary. However, the analysis model, a counterpart of the synthesis model, has not received much attention until recent years. The analysis model takes a different viewpoint to sparse representation, and it assumes that the product of an analysis dictionary and a signal is sparse. Compared with the synthesis model, this model tends to be more expressive to represent signals, as a much richer union of subspaces can be described. This talk focuses on the analysis model and aims to discuss the two main challenges: analysis dictionary learning (ADL) and signal reconstruction. In the ADL problem, the dictionary is learned from a set of training samples so that the signals can be represented sparsely based on the analysis model, thus offering the potential to fit the signals better than pre-defined dictionaries. Among the existing ADL algorithms, such as the well-known Analysis K-SVD, the dictionary atoms are updated sequentially. The first part of this talk presents two novel analysis dictionary learning algorithms to update the atoms simultaneously. In particular, by adapting Simultaneous Codeword Optimization (SimCO), an algorithm proposed originally for the synthesis model, the Analysis SimCO algorithm is developed for the analysis model. This algorithm assumes that the analysis dictionary contains unit l2-norm atoms and learns the dictionary by optimization on manifolds. This framework allows multiple dictionary atoms to be updated simultaneously in each iteration. However, similar to the existing ADL algorithms, the dictionary learned by Analysis SimCO may contain similar atoms. To address this issue, Incoherent Analysis SimCO is proposed by employing a coherence constraint and introducing a decorrelation step to enforce this constraint. The competitive performance of the proposed algorithms is demonstrated in the experiments for recovering synthetic dictionaries and removing additional noise in images, as compared with existing methods. The second part of this talk discusses how to reconstruct signals with learned dictionaries under the analysis model. This is demonstrated by a challenging application problem: multiplicative noise removal (MNR) of images. In the existing sparsity motivated methods, the MNR problem is addressed using pre-defined dictionaries, or learned dictionaries based on the synthesis model. In this talk, the potential of analysis dictionary learning for the MNR problem is discussed. Two new MNR algorithms using analysis dictionary learning are presented. In the first algorithm, a dictionary learned based on the analysis model is employed to form a regularization term, which can preserve image details while removing multiplicative noise. In the second algorithm, a smoothness regularizer is introduced to the reconstruction formulation to improve the recovery quality of smooth areas in images. This regularizer can be seen as an enhanced Total Variation (TV) term with an additional parameter controlling the level of smoothness. To address the optimization problem of this model, the Alternating Direction Method of Multipliers (ADMM) is adapted and a relaxation technique is developed to allow variables to be updated flexibly. Experimental results show the superior performance of the proposed algorithms as compared with three recently proposed algorithms for a range of noise levels.

	Prof. Wenying Feng Department of Computing & Information Systems and Department of Mathematics, Trent University, Ontario, Canada
Biography: Dr. Wenying Feng is a Professor in the Department of Computing & Information Systems and Department of Mathematics at Trent University, Ontario, Canada, and an Adjunct Professor at the School of Computing, Queen's University, Canada. She received her Ph.D. in Mathematics from the University of Glasgow in 1997 and has published more than 90 articles in journals and conference proceedings covering areas of both Mathematics and Computer Science. She served as the director of the Applied Modeling and Quantitative Methods graduate program at Trent University (2006–2009) and a panelmember for the Early Researcher Awards (ERA) of the Ontario Ministry of Research and Innovation (2009–2011). Her current research interests include differential equations, nonlinear analysis, big data modeling and analytics, machine learning and computational algorithms. Dr. Feng has presented numerous invited talks, served as program chairs and organized special tracks for multiple international conferences. Her research has been supported by NSERC (Natural Sciences and Engineering Research Council of Canada) and MITACS (Mathematics of Information Technology and Complex Systems).She has been a reviewer for the American Mathematical Society since 2004 and a board member for the International Society of Computers and Their Applications (ISCA, 2013-2016). Currently she also serves as the Chair of the Department of Mathematics at Trent University.
Speech Title: Computational Modelling with Real World Applications Abstract: In this talk, I will present two computational models for machine learning and data mining with real-world applications. The first model is on deep learning neural network and is developed by introducing a new activation function that offers higher gradients and faster learning rate. Moreover, to simplify the network structure, a sparsity function is applied to the hidden layer of the network. A technique that swaps the activation functions of fully trained network to logistic function is introduced to enhance system performance. The second model applies data mining techniques and is applied to detect coalition attacks, which is one of the most common types ofattacks in the industry of online advertising. In development of new algorithms, we attempt to mitigatethe problem of frauds by proposing a hybrid framework that detects thecoalition attacks based on multiple metrics. We also articulate the theoreticalbasis for these metrics to be integrated into the hybrid framework. Furthermore,we instance the framework with two metrics and develop a detection system that identifies the coalition attacks from two distinguish perspectives.

More information about invited speakers will be updated soon.