Learning admissible heuristics while solving problems

A method is presented that causes A * to return high quality solutions while solving a set of problems using a non-admissible heuristic. The heuristic guiding the search changes as new information is learned during the search, and it converges to an admissible heuristic which 'contains the insight ' of the original nonadmissible one. After a finite number of problems, A * returns only optimal solutions. Experiments on sliding tile problems suggest that learning occurs very fast. Beginning with hundreds of randomly generated problems and an overestimating heuristic, the system learned sufficiently fast that only the first problem was solved non-optimally. As an application we show how one may construct heuristics for finding high quality solutions at lower cost than those returned by A * using available admissible heuristics.