Keynote Speech

Improving Metaheuristic Algorithms Using Information Feedback Model

In most metaheuristic algorithms, the individual update process does not (fully) utilize the individual information generated in previous iterations. If this useful information can be fully utilized in subsequent optimization processes, the quality of the feasible solutions produced by the algorithm will be greatly improved. Based on this, a method of reusing available information of previous individuals to guide subsequent search is proposed. In this method, the previous useful information is fed back to the individual update process, and then six information feedback models are proposed. In these models, the individuals of previous iterations are selected in a fixed or random way, and then the useful information of the selected individuals is applied to the individual update process. Then, based on the individuals generated and selected by the basic algorithm, a simple fitness weighting method is used to generate new individuals. Six different information feedback models are applied to 10 metaheuristic algorithms to generate new algorithms and verify the performance of the proposed information feedback model. Experiments show that these new algorithms are significantly better than the basic algorithms on 14 standard test functions and 10 CEC 2011 real world problems, and further prove the effectiveness of the proposed information feedback model. At the same time, the model is applied to solve many-objective optimization methods (MOEA/D and NSGA-III), and good results are achieved.

Gai-Ge Wang is an associate professor in Ocean University of China, China. His entire publications have been cited over 9000 times (Google Scholar). Fifteen and sixty-six papers are selected as Highly Cited Paper by Web of Science, and Scopus (till July, 2021), respectively. One paper is selected as “Top Articles from Outstanding S&T Journals of China-F5000 Frontrunner”. He was selected as one of “2020 Highly Cited Chinese Researchers” in computer science and technology by Elsevier. He was selected as World’s Top 2% Scientists 2020, ranked 3840 in single 2019 (ranked 30762 in 2017), and ranked 88554 in career-long citation impact. One of his papers was selected as “100 Most Influential International Academic Papers in China”, One of his paper ranks 1 in the selection of the latest high-impact publications in computer science by Chinese researchers across from Springer Nature in 2019. The latest Google h-index and i10-index are 52 and 103, respectively. He is senior member of SAISE, SCIEI, a member of IEEE, IEEE CIS, ISMOST. He served as Editorial Advisory Board Member of Communications in Computational and Applied Mathematics (CCAM), Associate Editor of IJCISIM, an Editorial Board Member of IEEE Access, Mathematics, IJBIC, Karbala International Journal of Modern Science, and Journal of Artificial Intelligence and Systems. He served as Guest Editor for many journals including Mathematics, IJBIC, FGCS, Memetic Computing and Operational Research. His research interests are swarm intelligence, evolutionary computation, and big data optimization.

Applying De-Clustering Concept to Information Hiding

Reversible steganography allows an original image to be completely restored after the extraction of hidden data embedded in a cover image. In this talk, I will talk about a reversible scheme based on declustering strategy for VQ compressed images. The declustering can be regarded as a preprocessing step to make the proposed steganographic method more efficient. The experimental results show that the time required for the embedding process in the proposed method is few. In addition, the reversible steganography allows an original image to be completely restored after the extraction of hidden data embedded in a cover image. In this talk, I will introduce a reversible scheme for VQ-compressed images that is based on a declustering strategy and takes advantage of the local spatial characteristics of the image. The main advantages of this method are ease of implementation, low computational demands,and no requirement for auxiliary data.

Professor Chang has worked on many different topics in information security, cryptography, multimedia image processing and published several hundreds of papers in international conferences and journals and 38 books. He was cited over 38,000 times and has an h-factor of 91 according to Google Scholar. Several well-known concepts and algorithms were adopted in textbooks. He also worked with the National Science Council, Ministry of Technology, Ministry of Education, Ministry of Transportation, Ministry of Economic Affairs and other Government agencies on more than 100 projects and holds 17 patents, including one in US and two in China.

He served as Honorary Professor, Consulting Professor, Distinguished Professor, Guest Professor at over 50 academic institutions and received Distinguished Alumni Award's from his Alma Master's. He also served as Editor or Chair of several international journals and conferences and had given almost a thousand invited talks at institutions including Chinese Academy of Sciences, Academia Sinica, Tokyo University, Kyoto University, National University of Singapore, Nanyang Technological University, The University of Hong Kong, National Taiwan University and Peking University.

Professor Chang has mentored 63 PhD students and 195 master students, most of whom hold academic positions at major national or international universities. He has been the Editor-in-Chief of Information Education, a magazine that aims at providing educational materials for middle-school teachers in computer science. He is a leader in the field of information security of Taiwan. He founded the Chinese Cryptography and Information Security Association, accelerating information security the application and development and consulting on the government policy. He is also the recipient of several awards, including the Top Citation Award from Pattern Recognition Letters, Outstanding Scholar Award from Journal of Systems and Software, and Ten Outstanding Young Men Award of Taiwan. He was elected as a Fellow of IEEE and lET in 1998, a Fellow of CS in 2020, and an AAIA Fellow in 2021 for his contribution in the area of information security.

Explainable AI for multimedia processing

Explainable AI (XAI) is essential if users want to understand, appropriately trust, and effectively manage this incoming generation of artificially intelligent partners. DARPA introduced this model in 2017. The problem of explainability is, to some extent, the result of AI’s success. In the early days of AI, the predominant reasoning methods were logical and symbolic. These early systems were reasoned by performing some form of logical inference on (somewhat) human readable symbols. Early systems could generate a trace of their inference steps, which then became the basis for explanation. XAI is needed for multimedia processing, models, learning paradigms, and architectures as tools for trusted results.

Vaclav Snasel's research and development experience include over 30 years in the Industry and Academia. He works in a multi-disciplinary environment involving artificial intelligence, information retrieval, nature and biologically inspired computing, data mining, and applied to various real-world problems. He studied numerical mathematics at Palacky University in Olomouc, a PhD degree obtained at Masaryk University in Brno. He teaches as a professor at VSB – Technical University of Ostrava. From 2001 to 2009, he worked as a researcher at The Institute of Computer Science of the Academy of Sciences of the Czech Republic. Since 2009 he works as head of the research program Knowledge management at IT4Innovation National Supercomputing Center; from 2010 until 2017, he works as dean of the Faculty of Electrical Engineering and Computer Science, and from 2017 he is rector of VSB-Technical University of Ostrava. He has given 18 plenary lectures and conference tutorials in these areas. He has authored/co-authored several refereed journal/conference papers and book chapters. He has published more than 600 papers (460 papers are indexed at Web of Science, 660 indexed at Scopus).

He has given 18 plenary lectures and conference tutorials in these areas. He has authored/co-authored several refereed journal/conference papers and book chapters. He has published more than 600 papers (460 papers are indexed at Web of Science, 660 indexed at Scopus).

Web of Science H-index 21, Citation index 2426, Scopus H-Index 27, Citation index 4350, Google Scholar H-Index 40, Citation index 8660.