Prioritized-LRTA*: Speeding Up Learning via Prioritized Updates

Modern computer games demand real-time simultaneous control of multiple agents. Learning real-time search, which interleaves planning and acting, allows agents to both learn from experience and respond quickly. Such algorithms re- quire no prior knowledge of the environment and can be deployed without pre-processing. We introduce Prioritized- LRTA*, an algorithm based on Prioritized Sweeping. P- LRTA* focuses learning on important areas of the search space, where 'importance' is determined by the magnitude of the updates made to neighboring states. Empirical tests on path-finding in commercial game maps show a substan- tial learning speed-up over state of the art learning real-time heuristic search algorithms.