Invited Speakers

---

Nicolai Petkov is professor of computer science with a chair in Intelligent Systems and Parallel Computing at the University of Groningen since 1991. He received his doctoral degree at the Dresden University of Technology in Germany. After graduation he worked at several universities and in 1991 he was appointed professor of Computer Science at the University of Groningen. He was the PhD thesis director of 31 scientists till now. He was scientific director of the Institute for Mathematics and Computer Science (now Bernoulli Institute) from 1998 to 2009, and currently he is vice-chairman of the University Council and chairman of the Science Faction since 2011. Petkov is associate editor of several scientific journals (e.g. J. Image and Vision Computing). He co-organized and co-chaired the 10th International Conference of Computer Analysis of Images and Patterns CAIP 2003 in Groningen, the 13th CAIP 2009 in Munster, Germany, the 16th CAIP 2015 in Valletta, Malta, the International Workshops Braincomp 2013, 2015 and 2017 on Brain-Inspired Computing in Cetraro, Italy, and the International Conference on Applications of Intelligent Systems APPIS 2018 and 2019 in Las Palmas de Gran Canaria, Spain. Petkov's research interests are in the field of development of pattern recognition and machine learning algorithms that he applies to various types of big data: image, video, audio, text, genetic, phenotype, medical, sensor, financial, web, etc.[2] He develops methods for the generation of intelligent programs that are automatically configured using training examples of events and patterns of interest. See further www.cs.rug.nl/is

Title: Representation learning with trainable COSFIRE filters

Keywords: Representation learning, deep learning, COSFIRE trainable filters

Abstract

In order to be effective, traditional pattern recognition methods typically require a careful manual design of features, involving considerable domain knowledge and effort by experts. The recent popularity of deep learning is largely due to the automatic configuration of effective early and intermediate representations of the data presented. The downside of deep learning is that it requires a huge number of training examples.

Trainable COSFIRE filters are an alternative to deep networks for the extraction of effective representations of data. COSFIRE stands for Combinations of Shifted Filter Responses. Their design was inspired by the function of certain shape selective neurons in areas V4 and TEO of visual cortex. A COSFIE filter is configured by the automatic analysis of a single pattern. The highly non-linear filter response is computed as a combination of the responses of simpler filters, such as Difference of (color) Gaussians or Gabor filters, taken at different positions of the concerned pattern. The identification of the parameters of the simpler filters that are needed and the positions at which their responses are taken is done automatically. An advantage of this approach is its ease of use as it requires no programming effort and little computation - the parameters of a filter are derived automatically from a single training pattern. Hence, a large number of such filters can be configured effortlessly and selected responses can be arranged in feature vectors that are fed into a traditional classifier.

This approach is illustrated by the automatic configuration of COSFIRE filters that respond to randomly selected parts of many handwritten digits. We configure automatically up to 5000 such filters and use their maximum responses to a given image of a handwritten digit to form a feature vector that is fed to a classifier. The COSFIRE approach is further illustrated by the detection and identification of traffic signs and of sounds of interest in audio signals.

The COSFIRE approach to representation learning and classification yields performance results that are comparable to the best results obtained with deep networks but at a much smaller computational effort. Notably, COSFIRE representations can be obtained using numbers of training examples that are many orders of magnitude smaller than those used by deep networks.

---

---

Prof. Dr. Satoshi Tojo received B.E., M.E. and Ph.D. from the University of Tokyo. After he worked at Mitsubishi Research Institute, Inc. in 1983-1995, he joined Japan Advanced Institute of Science and Technology (JAIST) as an associate professor of School of Information Science in 1995, and a full professor from 2000 to this day. His interest includes formal semantics of natural language, logic of knowledge and belief in agent communication. His recent works concern also the iterated learning of grammar acquisition, as well as grammatical analysis of music.

Title: Music and AI

Keywords: Music, Natural language, Artificial intelligence, Creativity.

Abstract

The biological origins of music and language are said to be one the same as we use ears to hear and throats to sing or to utter, however, only we human beings possess both. In this talk, we discuss the applicability of techniques on natural language processing for the music, because there must be grammatical structure either in music. Within this talk, we will make two important digressions. One is what is the communication. Language seems a meaningful, communicative part of the common origin; then, what is the meaning of music? The other digression is the cognitive bias of infant's language acquisition, which suggests us the possibility of statistical learning of music knowledge. Thereafter, we present our formalization of music in syntactic tree, our implementation of music analyzer and several application systems, including the similarity assessing and the morphing. Finally, we close the talk by the open discussion on whether artificial intelligence come to obtain musical aesthetics and to compose music.

---

---

Chidchanok Lursinsap received the B.Eng. degree (honors) in computer engineering from Chulalongkorn University, Bangkok, Thailand, in 1978 and the M.S. and Ph.D. degrees in computer science from the University of Illinois at Urbana-Champaign, Urbana, in 1982 and 1986, respectively. He was a Lecturer at the Department of Computer Engineering, Chulalongkorn University, in 1979. In 1986, he was a Visiting Assistant Professor at the Department of Computer Science, University of Illinois at Urbana-Champaign. From 1987 to 1996, he worked at The Center for Advanced Computer Studies, University of Louisiana at Lafayette, as an Assistant and Associate Professor. After that, he came back to Thailand to establish Ph.D. program in computer science at Chulalongkorn University and became a Full Professor. His major research interests include neural learning and its applications to other science and engineering areas.

Title: Learning with Arbitrary Class Drift in Streaming Data and Applications

Keywords: TBA

Abstract

Arbitrary class drift is a situation where the classes of incoming data arbitrarily change with unknown probability and time periods. The same data already learned may enter the learning process with new classes. Furthermore, the amount of data flow into the learning process may overflow the memory. This scenario actually occurs in various problems. The preliminary solution to this problem consisting of a new neural network structure and learning concept will be discussed.

---

---

Dr. Sumate Charoenchaidet received B.Sc., M.Sc. and Ph.D. from the Chulalongkorn University in 1995, 1997 and 2002, respectively. He also got his Ph.D. degree, as a dual degree, from the University of Michigan, Ann Arbor, USA. After he worked at Thai Polyethylene for eleven years as catalyst researcher, manager and Senior research in 2002-2013, he joined SCG Chemicals Co., Ltd as a senior researcher and the group head of polyolefin central research. His interest includes borate co-catalyst, ethylene polymerization and copolymerization.

Title: Recent Research and Development of Polyolefins Technology in SCG-Chemicals

Keywords: Polyolefin, Process technology, Polyolefin catalysis, ROTO molding, Polyolefin wax, Excellent packaging

Abstract

SCG-Chemicals has launched Polyolefin business since 1983 in Mapthaphud industrial estate, Rayong province. Now a day, we have expanded to upstream and many various downstream-products, for example, composited materials, PTA, MMA etc. Therefore, our vision has been intensified to Differentiated products and High value added products in the future portfolio. Research and Development program has been established since 2002 as Technology Building Blocks for Process technology and Polyolefin catalysis, Polymer Structure and Processing in 2006 and Compounding technology in 2007. Our R&D mission is to strengthen SCG Chemicals to be market leader in South-East of Asia. Recently, we have provided premium pipe applications, ROTO molding, polyolefin wax, excellent packaging etc. to the market by in-house R&D task forces. We are thriving to create new and sustainable competitive advantages by delivering innovations that matters most to the business and leveraging the uses of IP, knowing why we are here and what do we need to be specialized. Finally, we would be appreciative to invite specialists and researches from around the world to join in to create value together. As SCG's current promising is "Passion for Better" for challenging to change and offer better solution.

---