Real-time tree search with pessimistic scenarios: Winning the NeurIPS 2018 Pommerman Competition

Takayuki Osogami, Toshihiro Takahashi
Proceedings of The Eleventh Asian Conference on Machine Learning, PMLR 101:583-598, 2019.

Abstract

Autonomous agents need to make decisions in a sequential manner, under partially observable environment, and in consideration of how other agents behave. In critical situations, such decisions need to be made in real time for example to avoid collisions and recover to safe conditions. We propose a technique of tree search where a deterministic and pessimistic scenario is used after a specified depth. Because there is no branching with the deterministic scenario, the proposed technique allows us to take into account the events that can occur far ahead in the future. The effectiveness of the proposed technique is demonstrated in Pommerman, a multi-agent environment used in a NeurIPS 2018 competition, where the agents that implement the proposed technique have won the first and third places.

Cite this Paper

BibTeX
@InProceedings{pmlr-v101-osogami19a, title = {Real-time tree search with pessimistic scenarios: Winning the NeurIPS 2018 Pommerman Competition}, author = {Osogami, Takayuki and Takahashi, Toshihiro}, booktitle = {Proceedings of The Eleventh Asian Conference on Machine Learning}, pages = {583--598}, year = {2019}, editor = {Lee, Wee Sun and Suzuki, Taiji}, volume = {101}, series = {Proceedings of Machine Learning Research}, month = {17--19 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v101/osogami19a/osogami19a.pdf}, url = {https://proceedings.mlr.press/v101/osogami19a.html}, abstract = {Autonomous agents need to make decisions in a sequential manner, under partially observable environment, and in consideration of how other agents behave. In critical situations, such decisions need to be made in real time for example to avoid collisions and recover to safe conditions. We propose a technique of tree search where a deterministic and pessimistic scenario is used after a specified depth. Because there is no branching with the deterministic scenario, the proposed technique allows us to take into account the events that can occur far ahead in the future. The effectiveness of the proposed technique is demonstrated in Pommerman, a multi-agent environment used in a NeurIPS 2018 competition, where the agents that implement the proposed technique have won the first and third places.} }
Endnote
%0 Conference Paper %T Real-time tree search with pessimistic scenarios: Winning the NeurIPS 2018 Pommerman Competition %A Takayuki Osogami %A Toshihiro Takahashi %B Proceedings of The Eleventh Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2019 %E Wee Sun Lee %E Taiji Suzuki %F pmlr-v101-osogami19a %I PMLR %P 583--598 %U https://proceedings.mlr.press/v101/osogami19a.html %V 101 %X Autonomous agents need to make decisions in a sequential manner, under partially observable environment, and in consideration of how other agents behave. In critical situations, such decisions need to be made in real time for example to avoid collisions and recover to safe conditions. We propose a technique of tree search where a deterministic and pessimistic scenario is used after a specified depth. Because there is no branching with the deterministic scenario, the proposed technique allows us to take into account the events that can occur far ahead in the future. The effectiveness of the proposed technique is demonstrated in Pommerman, a multi-agent environment used in a NeurIPS 2018 competition, where the agents that implement the proposed technique have won the first and third places.
APA
Osogami, T. & Takahashi, T.. (2019). Real-time tree search with pessimistic scenarios: Winning the NeurIPS 2018 Pommerman Competition. Proceedings of The Eleventh Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 101:583-598 Available from https://proceedings.mlr.press/v101/osogami19a.html.

Related Material