The 3rd Asian Conference on Artificial Intelligence Technology
July 5-7 2019, Chongqing, China
Keynote & Invited Speeches
Many fields, like physics, neuroscience, chemistry, and sociology, investigate phenomena by processing multivariate measurementsadvantageously represented as a sequence of attributed graphs. Graphs come in different forms, with variable attributes, topology, and ordering, making it difficult to perform a mathematical analysis in the graph space. Within this framework, we are interested in processing graph datastreams to solve applications e.g., detect structural changes in the graphsequence, a situation associated with time variance, faults, anomalies or events of interestas well as design sophisticated processing like those requested by predictors.
On the change detection front, theoretic results show that, under mild hypotheses, the confidence level of an event detected in the graph domain can be associated with another confidence level inan embedding space; this enables the identification of events in the graph domain by investigating embedded data. The opposite holds. However, evaluation of distances between graphs and identification of an appropriate embedding for the problem at hand are far from being trivial tasks with deep adversarial learning approaches and constant curvature manifold transformation showing to be appropriate transformations able to solve the problem. Deep autoregressive predictive models can then be designed to operate directly on graphs, hence providing the building blocks for other future sophisticated neural processing.
CESARE ALIPPI received the degree in electronic engineering cum laude in 1990 and the PhD in 1995 from Politecnico di Milano, Italy. Currently, he is a Professor with the Politecnico di Milano, Milano, Italy and Università della Svizzera italiana, Lugano, Switzerland. He is a visiting professor at the University of Kobe, Japan, the University of Guangzhou, China and Consultant Professor at the Northwestern Polytechnic in Xi’An, China. He has been a visiting researcher at UCL (UK), MIT (USA), ESPCI (F), CASIA (RC), A*STAR (SIN).
Alippi is an IEEE Fellow, Member of the Administrative Committee of the IEEE Computational Intelligence Society, Board of Governors member of the International Neural Network Society, Board of Directors member of the European Neural Network Society, Past Vice-President education of the IEEE Computational Intelligence Society, past associate editor of the IEEE Transactions on Emerging topics in computational intelligence, the IEEE Computational Intelligence Magazine, the IEEE-Transactions on Instrumentation and Measurements, the IEEE-Transactions on Neural Networks.
In 2018 he received IEEE CIS Outstanding Computational Intelligence Magazine Award, the 2016 Gabor award from the International Neural Networks Society and the IEEE Computational Intelligence Society Outstanding Transactions on Neural Networks and Learning Systems Paper Award; in 2013 the IBM Faculty award; in 2004 the IEEE Instrumentation and Measurement Society Young Engineer Award.
Current research activity addresses adaptation and learning in non-stationary environments, graph learning and Intelligence for embedded, IoT and cyber-physical systems.
He holds 8 patents, has published one monograph book, 7 edited books and about 200 papers in international journals and conference proceedings.
Evolutionary Computation (EC) encapsulates a class of stochastic optimisation algorithms, which are inspired by principles from natural and biological evolution. EC has been widely used for optimisation problems in many fields. Traditionally, EC methods have been applied for solving static problems. However, many real world problems are dynamic optimisation problems (DOPs), which are subject to changes over time due to many factors. DOPs have attracted a growing interest from the EC community in recent years due to the importance in the real-world applications of EC. This talk will first briefly introduce the concept of EC and DOPs, then review the main approaches developed to enhance EC methods for DOPs, and describe several detailed approaches developed for EC methods for DOPs. Finally, some conclusions will be made based on the work presented and the future work on EC for DOPs will be briefly discussed.
Shengxiang Yang is now a Professor in Computational Intelligence and Director of the Centre for Computational Intelligence (CCI), School of Computer Science and Informatics, De Montfort University. Before joining De Montfort University in July 2012, he worked as a Senior Lecturer in Brunel University from July 2010 to June 2012, as a Lecturer in the Universitu of Leicester from November 2000 to June 2010, respectively.
His research interest covers the following areas: genetic and evolutionary computation, memetic computing, swarm intelligence (particle swarm optimization and ant colony optimization), estimation of distributions algorithms, meta-heuristics and hyper-heuristics, artificial neural networks for scheduling, computational intelligence for combinatorial, multi-objective, and dynamic optimization problems, intelligent systems, bioinformatics, data mining and intelligent data analysis, network flow problems and algorithms, and relevant real-world applications.
He is a Senior Member of the Institute of Electrical and Electronics Engineers (IEEE). He was the Founding Chair of the IEEE CIS ISATC Task Force on Intelligent Network Systems (TF-INS, 2012-2017) and the Chair of the IEEE CIS ECTC Task Force on Evolutionary Computation in Dynamic and Uncertain Environments (ECiDUE, 2011-2017). He is in ProfPalsberg's list of about 1,000 computer scientists with the highest H-index (widely used index of research impact). He is a Member of the Peer Review College for the Engineering and Physical Sciences Research Council, UK.
In the current era of information overload, most of the datasets are high dimensional and often consist of many irrelevant and redundant features, thus making it complex and less efficient to model and analyze the data. It is, therefore, necessary to reduce the number of features required to achieve high classification performance. The feature selection algorithm looks for the optimal or most informative features by putting aside the redundant and irrelevant features, retaining accurate information and data structures where possible, resulting in efficient and more accurate predictive models. Various feature selection techniques have been developed and are continually nurtured to meet the evolving needs. This talk is intended to explore and analyze the effectiveness of Monte Carlo Tree Search (MCTS) in feature selection. MCTS is a dynamic search strategy and have proven its effectiveness in gaming domains with huge search spaces. MCTS finds the optimal solutions probabilistically by using lightweight random simulations. The feature selection task is modeled with a binary tree and modified MCTS is applied to the tree to select effective feature subsets. The proposed algorithms efficiently search the feature space and find the best feature subset within a few iterations to achieve high classification accuracy and reduced dimensions.
Jee-Hyong Lee is a Professor at Sungkyunkwan Univ., Korea. He is the vice-chair of Sungkyunkwan Convergence Institute for Information and Intelligence, the chair of Special Interest Group on Artificial Intelligence, Artificial Intelligence Society of Korea Institute of Information Scientists and Engineers, and an associate editor of International Journal of Fuzzy Logic and Intelligent Systems. He has performed research on data driven intelligent systems more than 15 years related with Artificial Intelligence, Machine Learning, Deep Learning, Data Mining, Text Mining and Big Data. Especially, he has focused on finding interesting and useful patterns in data and applying those patterns to solve problems in real world such as semi-conductor defect detection, hazardous weather prediction, recommendation, text summarization, etc.
In this talk, recent industrial applications of AR technology will be surveyedafter reviewing the smart glasses for industrial augmented reality (AR) in practice. As a case study of industrial AR, aircraft maintenance with AR has been researched for many years as a research project and is now being commercialized with newsmart glasses by Augmented Knowledge Corp. This initial research project was started collaboratively between Inha University and University of Southern Californiain 2008, which is funded by Korean Air and Airbus.
Aircraft maintenance and training play one of the most importantroles and symbolicapplications of digital transformation in the fourthindustrial revolutions as well as for digital twin paradigm.The maintenance process usually involves massive numbersof componentsand substantial knowledge of maintenance procedures. Maintenance tasks require technicians to follow rigorous proceduresto prevent operational errors in the maintenance process. In addition, the maintenance time is a cost-sensitive issue for airline carriers. In this research, intelligent augmented reality (IAR) system is implemented to minimize operation errors and time-related costs. This system alsohelp aircraft technicians cope with complex tasks by using an intuitive interface such as gestures, eye tracking and voice recognition fortheir maintenance tasks. A unified platform between computer vision and knowledge was invented for integrating computer vision with knowledge. Knowledge is introduced to deals with the various manuals which should be updated periodically. Overall testing of the IAR system is conducted at Korea Air Lines (KAL)hangars. According to the benchmark, theIAR system can help technicians to be more effective and accurate in performing their maintenance tasks, reduces the maintenance time about 30%.IAR-MAP which is a commercial version of IAR system provides a AR platform for B2B, which can integrate AR contents with the traditional legacy systems such as ERP and MRO.
Geun-Sik Jo is a Professor in Computer Science and Engineering at Inha University in Korea. He is also CEO and Founder of Augmented Knowledge Corp.In 1991, he completed his Ph.D. degreein Computer Science from the Graduate Center of City University of New York. He has been the General Chair and/or Technical Program Chair of numerous international conferences and workshops in artificial intelligence, knowledge management, and semantic webs. His research interests include computer vision, deep learning, augmented reality, semantic web, recommendation systems, constraint-directed scheduling and decision support systems. He served as a chairman of the Korea Intelligent Information Systems Society and also as a chair of the special interest group in Korea artificial intelligence society. He has numerous international patents, authored and coauthored five books and more than 300 technical papers published in journals and conference proceedings as well as ‘Innovative Applications of Artificial Intelligence” award from AAAI.
Dialog agent, which is also known as Chatbot, is recently attracting much attention from industrial fields and academic fields. The dialog agent takes a text or speech from users as an input, and gives a response text or synthesized voice. This provides a new way of interaction (e.g., voice or natural language text), and it can be applied to various industrial products (e.g., speaker, TV, search system). The change of interaction way is now reaching to its next step incorporating visual factors.
Mixed reality (MR) is a mixture concept of Augmented reality (AR) and Virtual reality (VR), and it allows the users to look at and interact with the visual agents. We design a dialog agent in the MR environment. Although the dialog agent is a virtual object, it has a indoor position in the real-world; we may interact with the agent when we are close enough to the agent. We believe that the dialog agent in MR environment has much potential to various industrial fields such as amusement park, education services, and healthcare services.
In 2010, he received a BS in Computer Science from Hanyang University, Korea, in 2012, an MSc in Computer Science from KAIST, Korea, and in 2016, a PhD in School of Computing from KAIST, Korea. He joined the faculty of the department of Bigdata Engineering at Soonchunhyang university, Asan city, Korea, in 2017.
His current research topics are text mining, information extraction, action recognition, and dialog systems, where hisfovorite techniques are topic modeling and deep learning.