Agent Mining: The Synergy of Agents and Data Mining
IEEE Intelligent Systems
Transfer learning in real-time strategy games using hybrid CBR/RL
IJCAI'07 Proceedings of the 20th international joint conference on Artifical intelligence
A data mining approach to strategy prediction
CIG'09 Proceedings of the 5th international conference on Computational Intelligence and Games
| Hi-index | 0.00 |
In real-time strategy games, resource gathering is a crucial part of constructing an army and becoming victorious. In this paper we present an algorithm for resource gathering and show how accumulated game data can be used to approximate travel times in a real-time strategy game. The algorithm builds upon a queueing system for resource collecting agents and optimises resource gathering by utilising travel times of agents in the game world. We implement the algorithm in the testbed of StarCraft: Brood War and compare it with the built-in method for resource gathering in this game. Experimental results show a gain in the amount of resources gathered when the algorithm is compared to the built-in method. In addition, the results demonstrate better predictability when our approach is used to gather resources for this particular game.