日時：7月25日（月）16:30-17:30
場所：東京工業大学　大岡山キャンパス　西８号館Ｗ棟１０Ｆ　Ｗ１００８

Speaker：Dr. Akihiro Kishimoto (IBM Research, Ireland)

Title: Efficient AND/OR search algorithms for exact MAP inference
       task over graphical models
Abstract
Graphical models provide a powerful framework for reasoning with
probabilistic information. Combinatorial maximization, or maximum
a posteriori (MAP) tasks arise in many applications and often can
be efficiently solved by search schemes, especially in the context
of AND/OR search spaces that are sensitive to the underlying
problem structure.
In this talk, I present the power of limited memory best-first
search over AND/OR search spaces, named RBFAOO, which performs
exact MAP inference over graphical models. I also present a
parallelized version of RBFAOO which runs in a shared-memory
environment. I show that RBFAOO is empirically superior to the
current state-of-the-art approaches based on AND/OR search,
especially on very hard problem instances.

ホスト　渡辺治（Ｃ０１）
watanabe@c.titech.ac.jp