Patrik Haslum

Note: I have moved. I do, however, occasionally read email sent to the address below. And it is not impossible that I may move back some day. ;)

Member of the Knowledge Processing Lab (KPLAB) in the Division for Artificial Intelligence and Integrated Computer Systems of the Department of Computer and Information Science at Linköping University.

mail: pahas AT ida DOT liu DOT se

Projects

WITAS

The project.

Heuristic Search Planning^*

Plan (and schedule) optimally by heuristic search.

Planning Domain Reduction

Removing redundancy in planning domains for improved efficiency.

Stocplan

An approach to planning and verification (model checking) by random walk.

Non-Deterministic Action Planning

Planning by regression in a logic of action and change.

Teaching

TDDB83 Logik

Logik för D1.

Modern AI Planning

Graduate course on AI Planning.

Publications

• Patrik Haslum, "Improving Heuristics Through Relaxed Search - An Analysis of TP4 and HSP*a in the 2004 Planning Competition". Journal of AI Research, vol 25, p. 233-267, 2006 (JAIR on-line copy).
• Patrik Haslum, Blai Bonet, Hector Geffner, "New Admissible Heuristics for Domain-Independent Planning". In Proc. AAAI, 2005 (PDF).
• Patrik Haslum, "Improving Heuristics Through Search" (poster/short paper). In Proc. European Conference on AI, 2004 (postscript, software).
• Patrik Haslum, "Patterns in Reactive Programs". In Proc. of the Cognitive Robotics Workshop, 2004. (postscript).
• Patrik Haslum and Ulrich Scholz, "Domain Knowledge in Planning: Representation and Use". In Proc. ICAPS 2003 Workshop on PDDL. (postscript).
• Patrik Haslum, "Partial State Progression: An Extension to the Bacchus-Kabanza Algorithm, with Applications to Prediction and MITL Consistency". In Proc. AIPS 2002 workshop on Planning via Model Checking (postscript).
• Patrik Haslum and Hector Geffner, "Heuristic Planning with Time and Resources". In Proc. 6th European Conference on Planning, 2001. To appear in Lecture Notes in Artificial Intelligence, Springer Verlag (postscript, software, related links).
• Patrik Haslum, "Models for Prediction". In Proc. IJCAI 2001 workshop on Planning under Uncertainty (postscript).
• Patrik Haslum and Hector Geffner, "Admissible Heuristics for Optimal Planning". In Proc. 5th International Conference on Artificial Intelligence Planning and Scheduling, 2000, p. 140 - 149. AAAI Press (postscript, software).
  • Note. There is an error in the paper: the condition for commutativity of actions (section "Commutativity Pruning") must also include that neither action adds a precondition of the other. Thus, commutativity is not the same as Graphplan-style "non-interference".
• Patrik Haslum and Peter Jonsson, "Planning with Reduced Operator Sets". In Proc. 5th International Conference on Artificial Intelligence Planning and Scheduling, 2000, p. 150 - 158. AAAI Press (postscript, software).
• Peter Jonsson, Patrik Haslum and Christer Bäckström, "Towards Efficient Universal Planning -- A Randomized Approach". Artificial Intelligence 117(1):1-29, 2000. (postscript, software).
• Patrik Haslum, "Model Checking by Random Walk". In Proc. 1999 ECSEL Workshop. (postscript, software).
• Patrik Haslum and Peter Jonsson, "Some Results on the Complexity of Planning with Incomplete Information". In Proc. of the 5th European Conference on Planning, Lecture Notes in Artificial Intelligence vol. 1809, 1999, p. 308 - 318. Springer Verlag (preprint version, postscript; final version available from Springer Link).
• Patrik Haslum, "An Investigation into the Computational Complexity of Planning in Non-Deterministic and Information-Incomplete Domains". Unpublished manuscript (postscript, 75K).

Thesis(es)

• "Admissible Heuristics for Automated Planning". Doctoral Thesis no. 1004, Linköping University. 2006. (available from LiU E-press).
• "Prediction as a Knowledge Representation Problem: A Case Study in Model Design". Licenciate Thesis no. 942, Linköping University. 2002. (postscript).

/P@trik Haslum, 2006.