IEEE International Conference on Development and Learning, 2010

Keynote Talks, Abstracts and Bios

 

Cognitive Development and the iCub Humanoid Robot


David Vernon

Italian Institute of Technology

Time: 7:00pm-8:00pm, Wednesday, August 18

Location: The Dahlmann Campus Inn

 

Abstract:

This talk addresses the central role played by development in cognition. The focus is on applying our knowledge of development in natural cognitive systems, i.e. human infants, to the problem of creating artificial cognitive systems in the guise of humanoid robots.

The work described in this talk is founded on the premise that (a) cognition is the process by which an autonomous self-governing agent acts effectively in the world in which it is embedded, that (b) the dual purpose of cognition is to increase the agent's repertoire of effective actions and its power to anticipate the need for and outcome of future actions, and that (c) development plays an essential role in the realization of these cognitive capabilities.

Our goal is to identify the key design principles for cognitive development. We do this by bringing together insights from four areas: enactive cognitive science, developmental psychology, neurophysiology, and computational modelling. We then discuss progress in applying these principles to the implementation of a cognitive architecture for the iCub, a open-systems humanoid robot which has been designed specifically as a common platform for research on embodied cognitive systems.

 

Bio Sketch:

David Vernon works as a freelance research scientist in the broad area of computer vision, robotics, and cognition. Recently, he has served as the coordinator of euCognition: the European Network for the Advancement of Artificial Cognitive Systems and as a member of the project team working on the creation of the iCub, an open-source cognitive humanoid robot. Over the past 32 years, he has held positions at Westinghouse Electric, Trinity College Dublin, the European Commission, the National University of Ireland, Maynooth, Science Foundation Ireland, University of Genoa, among others. He has authored five books on computer vision and has published over ninety papers in the fields of computer vision, robotics, and cognitive systems. His current research focus is on enactive approaches to cognition. He is a Senior Member of the IEEE, a Chartered Engineer of the Institution of Engineers of Ireland, and a past Fellow of Trinity College Dublin. He is an editor of the Springer series of Cognitive Systems Monographs (COSMOS).


The Developmental Discovery and Organization of Environmental Affordances


Rod Grupen

University of Massachusetts, Amherst

Time: 8:30am-9:30am, Thursday, August 19

Location: The Horace H. Rackham Building

 

Abstract:

Robots are clumsy and inflexible mostly because they are programmed by humans with little insight into what the world is like for a robot. Drawing on theories of how animals develop, I will argue for a different kind of machine that invents new states and actions in order to acquire skills for the real world, forms categories about controllable contexts, and learns about us and from us. Using UMass robots, I'll describe recent work that uses intrinsically motivated learning to discover controllable interactions with open, unstructured worlds.

 

Bio Sketch:

Professor Grupen has degrees in Physics, Mechanical Engineering, and received a Ph.D. in Computer Science from the University of Utah in 1988. He conducts research that integrates signal processing, control, dynamical systems, learning, and development as a means of designing controllers for intelligent systems. Grupen and his students develop hierarchical representations for sensorimotor policies that facilitate learning and transfer. This approach guides experimental research using dexterous mobility and manipulation platforms and has yielded techniques for motion control and collision avoidance, methods for coordinating multiple distributed robots, policies for dexterous grasping and walking gaits, and new concepts in the control of whole-body mobile manipulators. These tools have been used as a computational account of sensorimotor development in animals, including early childhood development, and new developmental programming techniques for intelligent machines that interact with open environments.

Professor Grupen is a Co-Editor-in-Chief of Robotics and Autonomous Systems, Editor of AIEDAM (AI in Engineering Design and Manufacture), and serves on several international program committees. He has been the Principal Investigator on grants from the NSF, DARPA, NASA, ONR, ARO, AFOSR, Microsoft, and iRobot. He co-founded the Embedded Systems instructional laboratory in which undergraduate and graduate students from many disciplines learn about building integrated sensorimotor systems. He received an Outstanding Teacher Award and recently received the Chancellor's medal and was designated a Distinguished Faculty Lecturer at the University of Massachusetts Amherst.


Core Capacities for Cooperation: Examples from Human Children and Chimpanzees


Felix Warneken

Harvard University

Time: 8:30am-9:30am, Friday, August 20

Location: The Horace H. Rackham Building

 

Abstract:

Cooperation is the hallmark of human social life, spanning from simple acts of helping another person to large-scale collaborative practices in which multiple people pool their efforts to achieve goals that lie beyond the means of any one individual. These behaviors raise interesting questions for cognitive science because they require specific social-cognitive capacities such as the ability to represent goal-directed action for helping and the formation of shared plans of actions for collaborative activities. In my talk, I will review studies from developmental and comparative psychology which aim at identifying the core capacities that (a) enable the most basic forms of cooperation in humans and our evolutionary relatives and (b) constitute the building blocks for later-emerging and more complex forms of cooperation in human development. These insights from developmental psychology might help to specify and test the cognitive architecture required for artificial agents to engage in diverse cooperative interactions with humans.

 

Bio Sketch:

Dr. Felix Warneken is Assistant Professor and Director of the Social Cognitive Development Group in the Psychology Department at Harvard University. Trained as a developmental and comparative psychologist, he conducts research with a focus on cooperation and social-cognitive development in children and great apes. He studied in Germany and the United States, receiving his doctoral degree from the Universität Leipzig while working at the Max Planck Institute for Evolutionary Anthropology (MPI EVA).  He continued as a postdoctoral researcher at MPI EVA and joined the Harvard Psychology Department in 2009. He received several awards and fellowships, including an Outstanding Dissertation Award by the Society for Research in Child Development in 2009 and a Novartis Fellowship in 2006. A study demonstrating altruistic helping in children and chimpanzees was named one of the 100 most important science stories in 2007 by Discover Magazine.


Sharks attack humans, but most sharks don't attack humans:  Learning to express generalizations
in language

 

Susan Gelman

University of Michigan

Time: 8:30am-9:30am, Saturday, August 21

Location: The Horace H. Rackham Building

Abstract:

My talk will examine the challenges that learners face when learning to express general categories (e.g., "sharks" in "Sharks attack humans").  These expressions, known as "generics", are both conceptually and linguistically challenging.  Although we experience the world in terms of individual objects and events, we must form abstractions that extend beyond these individual entities.  Moreover, the same forms of language that are used to express generics are also used to refer to particular individuals (compare "The dog is a 4-legged animal" to "The dog is sleeping").  I discuss different kinds of learning models, and suggest that generics are a default mode of generalization for human learners.

 

Bio Sketch:

Susan A. Gelman earned her Ph.D. in Psychology from Stanford University in 1984. She then joined the faculty of the University of Michigan, where she is currently the Frederick G. L. Huetwell Professor of Psychology. Her research interests focus on concept and language development in young children. She is interested in how children organize their experiences into categories, children's early theories of the world around them (particularly their emerging biological theories), children's "essentialist" beliefs, children's understanding of causality, and the role of language in expressing and conveying children's concepts. Dr. Gelman has published 6 books and monographs and over 100 articles and chapters, including The Essential Child (Oxford, 2003), which won the Eleanor E. Maccoby book prize of APA Division 7 and the 2005 book prize of the Cognitive Development Society. Professor Gelman received the APA Distinguished Scientific Award for Early Career Contribution to Psychology in the Developmental Area, the American Psychological Foundation Robert L. Fantz Award, the Chase Memorial Award, the Boyd McCandless Young Scientist Award from APA Division 7, and a J. S. Guggenheim Fellowship. She has served on the editorial boards of several journals, and is a fellow of the American Academy of Arts and Sciences.

 

Back