Search, Abstractions and Learning in Real-Time Strategy Games
| dc.contributor.author | Barriga, Nicolas A. | |
| dc.date.accessioned | 2021-04-23T09:30:28Z | |
| dc.date.available | 2021-04-23T09:30:28Z | |
| dc.date.issued | 2020 | |
| dc.description.abstract | Real-Time Strategy Games’ large state and action spaces pose a significant hurdle to traditional AI techniques. We propose decomposing the game into sub-problems and integrating the partial solutions into action scripts that can be used as abstract actions by a search or machine learning algorithm. The resulting high level algorithm produces sound strategic choices, and can then be combined with a low-level search algorithm to refine tactical choices. We show strong results in SparCraft, Starcraft: Brood War and $$\mu $$ μ RTS against state-of-the-art agents. We expect advances in RTS AI can be used in commercial videogames for playtesting and game balancing, while also having possible real-world applications. | de |
| dc.identifier.doi | 10.1007/s13218-019-00614-0 | |
| dc.identifier.pissn | 1610-1987 | |
| dc.identifier.uri | http://dx.doi.org/10.1007/s13218-019-00614-0 | |
| dc.identifier.uri | https://dl.gi.de/handle/20.500.12116/36275 | |
| dc.publisher | Springer | |
| dc.relation.ispartof | KI - Künstliche Intelligenz: Vol. 34, No. 1 | |
| dc.relation.ispartofseries | KI - Künstliche Intelligenz | |
| dc.subject | Deep convolutional neural networks | |
| dc.subject | Evolutionary algorithms | |
| dc.subject | Game tree search | |
| dc.subject | Real-time strategy games | |
| dc.title | Search, Abstractions and Learning in Real-Time Strategy Games | de |
| dc.type | Text/Journal Article | |
| gi.citation.endPage | 103 | |
| gi.citation.startPage | 101 |