Abstract | ||
---|---|---|
Case-based planners often face the problem of incurring more computational cost for retrieving a nd modifying a case for reuse, than what can be saved by reusing the case. We present a case-based planning system that learns the performance of a given plann er (called the default planner) in a training phase, a nd exploits this knowledge to retrieve and reuse cases such that planning effort is saved. The system does not involve any modification of the plan being reused. Furthermore, the system uses a very efficient metho d for matching a new problem with solved cases. The average-case performance of the system has been found to be significantly better than that of the default planner in a test domain. We hypothesize that this approach can be used to improve the performance of other planners as well. The effectiveness of the s ystem hinges mainly on the learning strategy and on extraction of the relevant features. |
Year | Venue | Keywords |
---|---|---|
2000 | AAAI/IAAI | case-based planning,planning performance,adaptive planner,feature extraction,learning |
Field | DocType | ISBN |
Scratch,Computer science,Reuse,Planner,Artificial intelligence,Machine learning | Conference | 0-262-51112-6 |
Citations | PageRank | References |
1 | 0.35 | 11 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Kreshna Gopal | 1 | 24 | 5.60 |
Thomas R. Ioerger | 2 | 623 | 59.10 |