Invited Speakers

ap

Kenneth O. Stanley
University of Central Florida

Carla Gomes
Cornell University

Maria Fasli
University of Essex

Ana Paiva
IST – Technical University of Lisbon

When Algorithms Inform Real Life: Novelty Search and the Myth of the Objective

Kenneth Stanley, University of Central Florida

Abstract
Have you ever had an algorithmic insight that changed how you thought about the world?  This talk will reflect on such a discovery, how it occurred, and why its implications extend well beyond the confines of AI and computer science.  The story begins with a surprising observation about the most successful users of an online picture-breeding service called Picbreeder, which leads to a paradoxical new algorithm called novelty search that searches without any objective.  However, the key insight behind this new algorithm, that the very act of forming an objective can block its own achievement, turns out more significant than just a step on the road to AI.  After all, almost all our efforts as human beings are framed by objectives.  Whether the aim is better algorithms, personal ambition, social change, or creative discovery, objectives loom behind almost every human endeavor.

The message of this talk: Maybe they shouldn’t.

Short bio
Kenneth O. Stanley is an associate professor in the Department of Electrical Engineering and Computer Science at the University of Central Florida. He received a B.S.E. from the University of Pennsylvania in 1997 and received a Ph.D. in 2004 from the University of Texas at Austin. He is an inventor of the Neuroevolution of Augmenting Topologies (NEAT), HyperNEAT, and novelty search algorithms for evolving complex artificial neural networks. His main research contributions are in neuroevolution (i.e. evolving neural networks), generative and developmental systems (GDS), coevolution, machine learning for video games, and interactive evolution. He has won best paper awards for his work on NEAT, NERO, NEAT Drummer, FSMC, HyperNEAT, novelty search, and Galactic Arms Race. He is an associate editor of IEEE Transactions on Computational Intelligence and AI in Games, on the editorial board of Evolutionary Computation journal, and on the ACM SIGEVO Executive Committee.  He is also a co-founder and the editor-in-chief of aigameresearch.org

Challenges for AI in Computational Sustainability

Carla Gomes, Cornell University

cgAbstract
Computational Sustainability is a new  interdisciplinary research field with the overall goal of developing computational models, methods, and tools to help manage the balance between environmental, economic, and societal needs for a sustainable future.  In this talk I will provide an overview of  Computational Sustainability, with examples ranging from wildlife conservation and biodiversity, to poverty mitigation, to large-scale deployment and management of renewable energy sources. I will  highlight overarching computational challenges for AI at the intersection of  constraint reasoning, optimization, machine learning, and dynamical systems. Finally I will discuss the need for a new approach that views computational sustainability problems as “natural” phenomena, amenable to a scientific methodology, in which principled experimentation, to explore problem parameter spaces and hidden problem structure, plays as prominent a role as formal analysis.

Short bio
Carla Gomes is a Professor of Computer Science at Cornell University, with joint appointments in the Dept. of Computer Science, Dept. of Information Science, and the Dyson School of Applied Economics and Management. Gomes’s central research themes  are the integration of concepts from constraint and logical reasoning, mathematical programming, and machine learning, for large scale combinatorial problems; the study of the impact of structure on problem hardness; and the use of randomization techniques to improve the performance of exact (complete) search methods. More recently, Gomes has become deeply immersed in research in the new field of Computational Sustainability.  Gomes is the Lead PI of an NSF Expeditions in Computing award on Computational Sustainability and the director of the newly established Institute for Computational Sustainability at Cornell University.  Gomes is a Fellow of the Association for the Advancement of Artificial Intelligence.

Exploring complex MAS systems: Applications & Challenges

Maria Fasli, University of Essex

Abstract
This talk will be presenting work on the modelling of complex socio-cognitive and economic multi-agent systems, some of their applications and challenges. In particular, work on the modelling of financial and electronic markets such as the foreign exchange market will be presented and insights acquired through the use of real and large datasets. Socio-cognitive aspects such as reputation and trust in dynamic environments such as markets will also be discussed.  Roles as the underlying concept for binding agents in social interactions will be explored. Finally, the challenges for the wider adoption of multi-agent systems as a design methodology and technology will be touched upon.

Short bio
Maria Fasli is a professor and the Head of the School of Computer Science and Electronic Engineering at the University of Essex. She obtained her Ph.D. in 2000 from Essex for her work on formal theories of agents entitled “Commodious Logics of Agents”. Her main research interests lie in formal theories of reasoning agents, and the applications of agent technology to e-commerce such as negotiation protocols, auctions, trading agents, supply chain management, as well as trust in electronic marketplaces. She has published papers in both areas. She has been leading teams in the International Trading Agent Competition (TAC) which features complex benchmark e-market problems since 2001 achieving consistently good results. In 2005, she was awarded a National Teaching Fellowship by the UK Higher Education Academy for her innovative approaches to learning and teaching. involving auctions.

Empathy in agents and robots

Ana Paiva, IST – Technical University of Lisbon

Abstract
During the last few years our social environment has been inundated with technologies that have revolutionized the way we live. Autonomous technological artifacts are now becoming a reality. Cars, vacuum cleaners, robots, or virtual characters can now act autonomously to support human activities. Yet, as these new technologies enter our social world, they must be able to interact with humans in a natural way.
This has opened up an exciting and challenging new research field on “social agents”. Social agents are autonomous software or robotic entities that interact socially with others, including humans. The aim is to get them to recognize others, their intentions, motivations and feelings. It entails perceiving and understanding affective states of others as well as acting accordingly. That is, social agents must have empathy.
In this talk I will provide an overview of this new challenging area of research, discussing the state-of-the-art in empathic, intelligent, autonomous virtual agents and robots. A concrete model for the creation of empathic agents will be presented, and illustrated with some examples of both virtual agents and social robots.

Short bio
Ana Maria Paiva is an Associate Professor in the Department of Computer Science and Engineering (Departamento de Engenharia Informática) of Instituto Superior Técnico from the Technical University of Lisbon ( Universidade Técnica de Lisboa). She is also the group leader of GAIPS, (Grupo de Agentes Inteligentes e Personagens Sintéticas), a research group on agents and synthetic characters at  INESC-ID.  Her main scientific interests lay in the area of Autonomous Agents, Embodied Conversational Agents and Robots and Multiagent Simulation Systems. Prof. Ana Paiva has been researching in the area of artificial intelligence for the past twenty years, having also taught at IST during that period.