The Taub Faculty of Computer Science Events and Talks

Classical AI planning is concerned with the following problem: Given a deterministic system, an initial system state, and a goal condition, find a sequence of actions that transforms the system from its initial state to a state that satisfies the goal condition. It was originally conceived in order to make robots autonomous, and has numerous applications. A simple and natural extension of classical planning is one where there are multiple agents, each with its own set of actions, that are trying to, cooperatively, transform the system into the goal state. In this talk I will discuss this model, two general techniques for solving it in a distributed setting, how these ideas can help us formulate algorithms that allow agents to cooperate while maintaining the privacy of certain information, and the multi-agent model can help us solve some single-agent problems efficiently.

Parts of this work are joint with Carmel Domshlak, Raz Nissim, Shlomi Maliah, and Guy Shani