AAAI 2006 workshop on
Learning for Search: schedule
8:45-9:00 Welcome and opening remarks, Wheeler and
Please note that oral presentations are scheduled to be 23 minutes, plus 7
minutes for questions.
9:00-10:30 Session 1: clause learning/constraint weighting
- 9:00 - 9:30 The Effect of Restarts on the Efficiency of Clause Learning.
Jinbo Huang. Presented by Toby Walsh since Jinbo can't attend due to visa
- 9:30 - 10:00 Learning from Failure in Constraint Satisfaction Search.
Diarmuid Grimes and Richard J.Wallace
- 10:00 - 10:30 Integrating Local Search and Complete Search with Restarts.
Carlos Ansotegui and Meinolf Sellman
10:30-11:00 Coffee break
11:00-12:30 Session 2: Run-time prediction/learning of heuristics
- 11:00 - 11:30 Estimating Search Tree Size. Philip Kilby, John Slaney,
Sylvie Thiebaux, and Toby Walsh
- 11:30 - 12:00 Performance Prediction and Automated Tuning of Randomized
and Parametric Algorithms: An Initial Investigation. Frank Hutter, Youssef
Hamadi, Holger H. Hoos, and Kevin Leyton-Brown
- 12:00 - 12:30 Toward Discriminative Learning of Planning Heuristics.
Yuehua Xu and Alan Fern
We are hoping to identify a nearby restaurant in advance and encourage all
participants to enjoy lunch together in informal conversation. If you are unable
to eat with the group, please be sure to return promptly for the afternoon sessions!
We now have a tentative reservation at LTK, the "Legal Test Kitchen", one block
from the WTC: http://www.ltkbarandkitchen.com/site/, 225 Northern Avenue.
When you face the Seaport Hotel coming from the WTC, it's one block to your
left. It's in the ground floor of a tower that has a sign on top saying "Now
Leasing, final phase" - you can't miss it. We should be there 12:40-1:50.
2:00-3:30 Session 3: Real-time search
3:30-3:35 Brief remarks, Wheeler and Frank
- 2:00 - 2:30 Prioritized-LRTA*: Speeding Up Learning via Prioritized Updates.
D. Chris Rayner and Katherine Davison and Vadim Bulitko and Jieshan Lu
- 2:00 - 3:00 PAC Reinforcement Learning Bounds for RTDP and Rand-RTDP.
Alexander L. Strehl, Lihong Li, and Michael L. Littman
- 3:00 - 3:30 Real-Time Adaptive A*. Sven Koenig and Maxim Likhachev
3:30-4:00 Coffee break
4:00-6:00 Poster session
(AAAI will provide 30 by 40 inch foamcore boards for the poster session.)
We gave oral presentations the option of also presenting a poster. The following
chose to do this:
- Storing learnt (no)goods in ROBDDs for solving structured CSPs.
Karim Boutaleb, Philippe Jegou and Cyril Terrioux
- Relative Support Weight Learning for Constraint Solving. Smiljana
Petrovic and Susan Epstein
- Discrepancy Search with Reactive Policies for Planning. Sungwook
- Replaying Types Sequences in Forward Heuristic Planning. Tomas de
la Rosa, Daniel Borrajo and Angel Garcia Olaya
- Some active learning schemes to acquire control knowledge for planning.
Raquel Fuentetaja and Daniel Borrajo
- Transfer of Learned Heuristics among Planners. Susana Fernandez,
Ricardo Aler and Daniel Borrajo
- Value Back-Propagation vs Backtracking in Real-Time Heuristic Search.
Sverrir Sigmundarson and Yngvi Bjoernsson
- State Abstraction for Real-time Moving Target Pursuit: A Pilot Study.
Vadim Bulitko and Nathan Sturtevant
- Lookahead Pathology in Real-Time Path-Finding. Mitja Lustrek and
- Directing a Portfolio with Learning. Mark Roberts and Adele E. Howe
- nLRTS: Improving Distance Vector Routing in Sensor Networks. Greg
Lee and Vadim Bulitko and Ioanis Nikolaidis
- An Empirical Evaluation of Automated Knowledge Discovery in a Complex
Domain. Jay H. Powell and John D. Hastings
- Integrating Local Search and Complete Search with Restarts. Carlos
Ansotegui and Meinolf Sellman
- Performance Prediction and Automated Tuning of Randomized and Parametric
Algorithms: An Initial Investigation. Frank Hutter, Youssef Hamadi, Holger
H. Hoos, and Kevin Leyton-Brown
- Toward Discriminative Learning of Planning Heuristics. Yuehua Xu
and Alan Fern
- Prioritized-LRTA*: Speeding Up Learning via Prioritized Updates D.
Chris Rayner and Katherine Davison and Vadim Bulitko and Jieshan Lu