ABSTRACT:

Although the field of heuristic search is over 50 years old, the last 6-7 years have seen numerous revisions to the foundational algorithms in the field. These include the theories for bidirectional search, for suboptimal search, and for improving the worst-case performance of fundamental algorithms such as A* and IDA*. This talk will give an overview of these new results, demonstrating the changes and their impact, many of which center around the notion of whether re-expansions are allowed during search.

BIO:

Nathan is a Fellow and Canada CIFAR AI Chair at Amii and a Professor in the Department of Computing Science at the University of Alberta. His research looks broadly at heuristic and combinatorial search problems, including both theoretical and applied approaches, with many applications in games. His work on pathfinding was used in the game Dragon Age: Origins and will appear in the upcoming Nightingale. Nathan”s work has won the best paper awards at the AAAI, and SoCS conferences, as well as the AI Journal Prominent Paper Award.

Hosted by Professor Richard Korf