Real-Time Heuristic Search with a Priority Queue
Abstract
Learning real-time search, which interleaves planning and acting, allows agents to learn from multiple trials and respond quickly. Such algorithms require no prior knowledge of the environment and can be deployed without pre-processing. We introduce Prioritized-LRTA* (P-LRTA*), a learning real-time search algorithm based on Prioritized Sweeping. P-LRTA* focuses learning on important areas of the search space, where the importance of a state is determined by the magnitude of the updates made to neighboring states. Empirical tests on path-planning in commercial game maps show a substantial learning speed-up over state-of-the-art real-time search algorithms. URL: http://www.cs.ualberta.ca/~rayner/IJCAI-RaynerD569.pdf
Cite
Text
Rayner et al. "Real-Time Heuristic Search with a Priority Queue." International Joint Conference on Artificial Intelligence, 2007.Markdown
[Rayner et al. "Real-Time Heuristic Search with a Priority Queue." International Joint Conference on Artificial Intelligence, 2007.](https://mlanthology.org/ijcai/2007/rayner2007ijcai-real/)BibTeX
@inproceedings{rayner2007ijcai-real,
title = {{Real-Time Heuristic Search with a Priority Queue}},
author = {Rayner, D. Chris and Davison, Katherine and Bulitko, Vadim and Anderson, Kenneth and Lu, Jieshan},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2007},
pages = {2372-2377},
url = {https://mlanthology.org/ijcai/2007/rayner2007ijcai-real/}
}