Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Learning for Search Papers from the AAAI Workshop Technical Report WS-06-11 AAAI Press American Association for Artificial Intelligence AAAI Press 445 Burgess Drive Menlo Park, California 94025 ISBN 978-1-57735-293-8 WS-06-11 Learning for Search Papers from the AAAI Workshop Wheeler Ruml and Frank Hutter, Cochairs Technical Report WS-06-11 AAAI Press Menlo Park, California Copyright © 2006, AAAI Press The American Association for Artificial Intelligence 445 Burgess Drive Menlo Park, California 94025 USA AAAI maintains compilation copyright for this technical report and retains the right of first refusal to any publication (including electronic distribution) arising from this AAAI event. Please do not make any inquiries or arrangements for hardcopy or electronic publication of all or part of the papers contained in these working notes without first exploring the options available through AAAI Press and AI Magazine (concurrent submission to AAAI and an another publisher is not acceptable). A signed release of this right by AAAI is required before publication by a third party. Distribution of this technical report by any means including electronic (including, but not limited to the posting of the papers on any Website) without permission is prohibited. ISBN 978-1-57735-293-8 WS-06-11 Manufactured in the United States of America Organizing Committee Wheeler Ruml, Palo Alto Research Center, USA Frank Hutter, University of British Columbia, Canada Program Committee Tom Carchrae, Cork Constraint Computation Center (4C), Ireland Susan Epstein, City University of New York, USA Hector Geffner, Universitat Pompeu Fabra (Barcelona), Spain Youssef Hamadi, MSR Cambridge, UK Frank Hutter (co-chair), University of British Columbia, Canada Henry Kautz, University of Washington, USA Sven Koenig, University of Southern California, USA Kevin Leyton-Brown, University of British Columbia, Canada Wheeler Ruml (co-chair), Palo Alto Research Center, USA Meinolf Sellmann, Brown University, USA Toby Walsh, University of New South Wales, Australia Shlomo Zilberstein, UMass Amherst, USA This AAAI–06 Workshop was held July 16, 2006, in Boston, Massachusetts USA iii Contents Preface / vii Oral Presentations The Effect of Restarts on the Efficiency of Clause Learning / 1 Jinbo Huang Learning from Failure in Constraint Satisfaction Search / 7 Diarmuid Grimes and Richard J.Wallace Disco — Novo — GoGo: Integrating Local Search and Complete Search with Restarts / 15 Meinolf Sellmann and Carlos Ansótegui Estimating Search Tree Size / 21 Philip Kilby, John Slaney, Sylvie Thiébaux, and Toby Walsh Performance Prediction and Automated Tuning of Randomized and Parametric Algorithms: An Initial Investigation / 28 Frank Hutter, Youssef Hamadi, Holger H. Hoos, and Kevin Leyton-Brown Toward Discriminative Learning of Planning Heuristics / 35 Yuehua Xu and Alan Fern Prioritized-LRTA*: Speeding Up Learning via Prioritized Updates / 43 D. Chris Rayner, Katherine Davison, Vadim Bulitko, and Jieshan Lu PAC Reinforcement Learning Bounds for RTDP and Rand-RTDP / 50 Alexander L. Strehl, Lihong Li, and Michael L. Littman Real-Time Adaptive A* / 57 Sven Koenig and Maxim Likhachev Poster Presentations Storing Learnt (no)goods in ROBDDs for Solving Structured CSPs / 65 Karim Boutaleb, Philippe Jégou, and Cyril Terrioux State Abstraction for Real-time Moving Target Pursuit: A Pilot Study / 72 Vadim Bulitko and Nathan Sturtevant Replaying Types Sequences in Forward Heuristic Planning / 80 Tomás de la Rosa, Daniel Borrajo, and Angel García Olaya v Transfer of Learned Heuristics among Planners / 85 Susana Fernández, Ricardo Aler, and Daniel Borrajo Some Active Learning Schemes to Acquire Control Knowledge for Planning / 93 Raquel Fuentetaja and Daniel Borrajo nLRTS: Improving Distance Vector Routing in Sensor Networks / 101 Greg Lee, Vadim Bulitko, and Ioanis Nikolaidis Lookahead Pathology in Real-Time Path-Finding / 108 Mitja Lustrek and Vadim Bulitko Relative Support Weight Learning for Constraint Solving / 115 Smiljana Petrovic and Susan Epstein An Empirical Evaluation of Automated Knowledge Discovery in a Complex Domain / 123 Jay H. Powell and John D. Hastings Directing a Portfolio with Learning / 129 Mark Roberts and Adele E. Howe Value Back-Propagation versus Backtracking in Real-Time Heuristic Search / 136 Sverrir Sigmundarson and Yngvi Björnsson Discrepancy Search with Reactive Policies for Planning / 142 Sungwook Yoon