talk

PLOW: A Collaborative Task Learning Agent

Get the Flash Player to see this video.

July 16, 2008

To be effective, an agent that collaborates with humans needs to be able to learn new tasks from humans they work with. This paper describes a system that learns executable task models from a single collaborative learning session consisting of demonstration, explanation and dialogue.

To accomplish this, the system integrates a range of AI technologies: deep natural language understanding, knowledge representation and reasoning, dialogue systems, planning/agent-based systems and machine learning. A formal evaluation shows the approach has great promise.

Speaker

James Allen

James Allen is a senior research scientist and associate director of the Institute for Human and Machine Cognition in Pensacola Florida. He also is the John H Dessauer Professor of Computer Science at the University of Rochester. He received his Phd in Computer Science from the University of Toronto and was a recipient of the Presidential Young Investigator award from NSF in 1984. A Founding Fellow of the American Association for Artificial Intelligence (AAAI), he was editor-in-chief of the journal Computational Linguistics from 1983-1993. Over the past twenty five years, he has been the principle investigator of research grants totaling over $30 million from agencies such as DARPA, ONR and NSF. He was general chair of the Second International conference on Principles of Knowledge Representation held in Boston in 1991, and the Fourth Int'l. Conference on AI Planning Systems in Pittsburgh in 1999.

He has authored numerous research papers in the areas natural language understanding, knowledge representation and reasoning, and spoken dialogue systems. He is the author of several books, including the influential textbook Natural Language Understanding, published by Benjamin Cummings in 1987, with a second edition published in 1995.

His research concerns defining computational models of intelligent collaborative and conversational agents, with a strong focus on the connection between knowledge representation and reasoning and language comprehension. He has published numerous research articles in the areas of natural language understanding, knowledge representation, temporal reasoning, planning and plan recognition. In recent years, he has been focussing on producing end-to-end working dialog systems that can connect everyday people to intelligent reasoning systems. In 2007, his paper on dialogue-based task learning won the best paper award at the National Conference on Artificial Intelligence (AAAI).

Institute for Human and Machine Cognition