Director, AI-ECON Research Center, Department of Economics National
Chengchi University, Taipei, Taiwan
Prof. Dr. Shu-Heng Chen is a Distinguished Professor in the Department of Economics, Dean of Office of International Cooperation, Director of the AI-ECON Research Center, and the organizer of Experimental Economics Laboratory at National Chengchi University. He also serves as the editor-in-chief of the Journal of New Mathematics and Natural Computation (World Scientific) and the Journal of Economic Interaction and Coordination (Springer), the associate editor of the Journal of Economic Behavior and Organization (Elsevier), the editor of International Journal of Financial Engineering and Risk Management as well as Economia Politica. Prof. Chen holds a Ph. D. in Economics from University of California at Los Angeles. He has more than 150 publications in international journals and edited book volumes. He is recognized as one of the founders and leaders in the field of agent-based computational economics and the first to introduce genetic programming into agent-based computational economics.
Title:Agent-Based Computational Economics and Experimental Economics: A Review of Some Recent Progresses
The recent advent of agent-based computational economics or econophysics (the applications of statistical physics to economics) accompanied by the 50-year-long experimental economics has made economics more like an experimental science, to be distinguished from the mainstream analytical economics and empirical economics. In this lecture, we will give a review of some recent progresses in the joint employment of agent-based computational economics (ACE) and experimental economics (EE) as two wheels of a wagon to move economic research forward. The idea is, within various familiar economic environments (experiments), to examine the role of individuals’ characteristics on their decision making, choices, and behavior, and its consequences. Among many possible individual characteristics, what particularly interest us at the current stage are cognitive capacity, personality, emotion, Identity, and gender. Specifically, we would like to demonstrate the use of this combined method (the two-wheel method) in our recent studies on the cognitive market experiments.
Research Director, UMR CNRS GÉOGRAPHIE-CITÉS, Paris, France.
Arnaud Banos obtained his Phd from the Université de Franche-Comté and his HDR degree from Université Paris 1 Sorbonne. He is a Research Director (DR2) at CNRS, Paris, France. His is currently the Director of CNRS lab “Géographie-Cités” from 2014 to 2019, and Director of the European research network “Spatial Simulation for Social Sciences” (http://s4parisgeo.cnrs.fr) since 2011. Previously, he was the Director of CNRS lab “Complex Systems Institute”, from 2012 to 2014.
Title: Social simulation of complex adaptive spatial systems
Most of the scientific objects privileged by geographers can be described as complex adaptive spatial systems. Such systems are composed of a large number of localised entities, interacting one with another through various networks of interaction, across different scales. From a single village to the global village, from a single urban street to the city on the move and to networks of cities, the range of scales mobilised is very wide. Therefore, not surprisingly, complexity sciences and their modelling and simulation tools constitute a main challenge for geography. At the very front in that new scientific battle are approaches allowing reproducing, by simulation, emergence of processes and structures in geographical space. Agent based simulations occupy a central place in the picture, thanks to the variety and flexibility of investigations they allow. Virtual labs can indeed be created, amongst which experimental approaches can be led. Once defined their characteristics and behaviours, agents are used to populate dynamic environments, in order to explore the possible conditions leading to the emergence of structures and processes. The goal of this communication is to present a tentative overview of the main challenges in the domain, based on several urban models simulating the emergence of cities, their spatial form and the mobility dynamics they host. Some of these models are advanced enough to be used in operational projects.
H. Van Dyke Parunak
Senior Scientist, Soar Technology, Ann Arbor, MI, USA
Van Parunak is a Senior Scientist at Soar Technology, where he leads applied research on the flow of information among data sources, heterogeneous models, and human analysts and decision-makers. His academic credentials include physics (AB, Princeton University, New Jersey, USA), Computer and Communication Sciences (MS, University of Michigan, Michigan, USA), archaeology and geography (MA, Jerusalem University College, Jerusalem, Israel), and ancient near eastern philology (MA and PhD, Harvard University, Cambridge, Massachusetts, USA). His research interests include self-organizing systems with emergent behavior; artificial intelligence, particularly knowledge representation and distributed planning and scheduling, in multiple domains including C2, information management, natural language processing, forecasting of dynamical systems, and tools for structured thinking and group work; and applications of statistical physics and probability theory to understanding the behavior of such systems. He has over 200 papers in journals and highly-refereed conferences, including numerous publications on social science simulation in MABS and CSSSA. He was a founding director of IFMAS, the predecessor of IFAAMAS, which sponsors the annual AAMAS conference, and has served repeatedly as sponsorship chair of AAMAS and as a member of the senior program committee. He co-chaired the 1992 DAI Workshop and two editions of MABS (2009, 2013), and is a member of the editorial boards of the Journal of Autonomous Agents and Multi-Agent Systems and the ACM Transactions on Autonomous and Adaptive Systems.
Title: From Statistical Mechanics to Human Society: Multi-Agent Systems and Social Self-Organization."
One of the most salient features of human society is its tendency to organize, to develop specialized roles and relations. Based on historical examples, students of society have proposed various theoretical explanations for how such structures arise and debated the consequences of alternative structures. But ethical and practical issues make it difficult to test these theories by conducting experiments on actual groups of people. This talk will outline main principles of self-organization, starting with the original insights developed in statistical mechanics, and including biological and ecological models as well as human society, as a guide to framing testable theories about social self-organization. It will then give examples, based on our own work and that of others, of how multi-agent simulation can be used to test these theories, and discuss methodological issues concerning the development, application, and validation of such models.
Wagner Meira Jr.
Full Professor, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil
INWeb - National Institute of Science and Technology for the Web
Wagner Meira Jr. obtained his PhD from the University of Rochester in 1997 and is Full Professor at the Computer Science Department at Universidade Federal de Minas Gerais, Brazil. He has published more than 200 papers in top venues and is co-author of the book Data Mining and Analysis - Fundamental Concepts and Algorithms published by Cambridge University Press in 2014. His research focuses on scalability and efficiency of large scale parallel and distributed systems, from massively parallel to Internet-based platforms, and on data mining algorithms, their parallelization, and application to areas such as information retrieval, bioinformatics, and e-governance.
Title: Mining the dynamics of online social networks for real-event prediction
The internet has been evolving from a communication media to an environment where users talk about the most diverse topics, reflecting the dynamics of the society at broad. Characterizing, understanding and modeling how the internet data may be used for assessing real events becomes a key component of many Internet-based applications and demands the development of new data mining models and techniques. Data mining in such scenarios is challenging because the data is intrinsically uncertain and multi-scale, the patterns to be mined are complex and evolve through time, and there is a huge amount of information that need to be processed in real time. In this talk we present a framework for the research and development of data mining models, algorithms and systems that target these challenging scenarios. We also present the Web Observatory, a platform for collecting, analyzing and presenting, at real time, information mined from social networks and the web, as well as some of its instances that focused on sports, politics, and health.