M
Michail G. Lagoudakis
Researcher at Technical University of Crete
Publications - 76
Citations - 4273
Michail G. Lagoudakis is an academic researcher from Technical University of Crete. The author has contributed to research in topics: Reinforcement learning & Robot. The author has an hindex of 24, co-authored 73 publications receiving 3970 citations. Previous affiliations of Michail G. Lagoudakis include Georgia Institute of Technology & University of Crete.
Papers
More filters
Journal ArticleDOI
Least-squares policy iteration
TL;DR: The new algorithm, least-squares policy iteration (LSPI), learns the state-action value function which allows for action selection without a model and for incremental policy improvement within a policy-iteration framework.
Proceedings Article
Coordinated Reinforcement Learning
TL;DR: These methods differ from many previous reinforcement learning approaches to multiagent coordination in that structured communication and coordination between agents appears at the core of both the learning algorithm and the execution architecture.
Proceedings ArticleDOI
Auction-Based Multi-Robot Routing
Michail G. Lagoudakis,Evangelos Markakis,David Kempe,Pinar Keskinocak,Anton J. Kleywegt,Sven Koenig,Craig A. Tovey,Adam Meyerson,Sonal Jain +8 more
TL;DR: A generic framework for auction-based multi-robot routing is suggested and auction methods are shown to offer theoretical guarantees for such a variety of bidding rules and team objectives.
Proceedings ArticleDOI
Simple auctions with performance guarantees for multi-robot task allocation
TL;DR: PRIM ALLOCATION is a simple and fast approximate algorithm for allocating targets to robots which provably computes allocations whose total cost is at most twice as large as the optimal total cost.
Proceedings Article
Reinforcement learning as classification: leveraging modern classifiers
TL;DR: It is argued that the use of SVMs, particularly in combination with the kernel trick, can make it easier to apply reinforcement learning as an "out-of-the-box" technique, without extensive feature engineering.