Learning to solve problems by searching for macro-operators
Learning to solve problems by searching for macro-operators
Depth-first iterative-deepening: an optimal admissible tree search
Artificial Intelligence
On the construction of heuristic functions
On the construction of heuristic functions
Machine learning: a theoretical approach
Machine learning: a theoretical approach
Improving path planning with learning
ML92 Proceedings of the ninth international workshop on Machine learning
Case-based reasoning
Robot Motion Planning
Learning Search Control Knowledge: An Explanation-Based Approach
Learning Search Control Knowledge: An Explanation-Based Approach
Hi-index | 0.00 |
We study a simple, general framework for search called bootstrap search, which is defined as global search using only a local search procedure along with some memory for learning intermediate subgoals. We present a simple algorithm for bootstrap search, and provide some initial theory on its performance. In our theoretical analysis, we develop a random digraph problem model and use it to make some performance predictions and comparisons. We also use it to provide some techniques for approximating the optimal resource bound on the local search to achieve the best global search. We validate our theoretical results with empirical demonstration on the 15-puzzle. We show how to reduce the cost of a global search by 2 orders of magnitude using bootstrap search. We also demonstrate a natural but not widely recognized connection between search costs and the lognormal distribution.