Winning Prediction in WoW Strategy Game Using Evolutionary Learning
Winning Prediction in WoW Strategy Game Using Evolutionary Learning
复制标题
使用进化学习在《魔兽世界》策略游戏中进行获胜预测
DOI：
10.1109/is3c.2014.191
复制
发表时间：
2014
期刊：
影响因子：
--
通讯作者：
Jyh
中科院分区：
文献类型：
--
作者：
Tain;Shao;Chiu;D. Tsaih;Jyh
在过去的几十年里，实时战略(RTS)游戏越来越受欢迎，并在视频游戏联盟中变得普遍。然而，对于创建人类级别的游戏AI来说，一个很大的挑战是对手的种族特征不同，他们在敌人单位的位置是部分可见的。为了克服这一限制，我们探索了基于进化的方法来估计遇到的敌人单位的位置。在本文中，我们提出了一种有效的框架来预测实时策略博弈中不同比赛之间的胜率。我们将状态估计表示为一个优化问题，并通过学习专业的星际游戏重放语料库来自动学习基于进化的模型的参数。基于进化的模型跟踪对手单位，并为激活我们的星际战靴中的战术行为提供条件。我们的结果表明，与基准方法相比，结合学习的基于进化的模型可以将EISBot的性能提高60%。
Over the past decades, real-time strategy (RTS) games have steadily gained in popularity and have become common in video game leagues. However, a big challenge for creating human-level game AI is the different traits of races of opponents and their locations of enemy units are partially observable. To overcome this limitation, we explore evolutionary-based approach for estimating the location of enemy units that have been encountered. In this paper, we propose an efficient framework to predict the winning ratio between the different races used in the real-time strategy game. We represent state estimation as an optimization problem, and automatically learn parameters for the evolutionary-based model by learning a corpus of expert Star Craft replays. The evolutionary-based model tracks opponent units and provides conditions for activating tactical behaviors in our Star Craft boot. Our results show that incorporating a learned evolutionary-based model improves the performance of EISBot by 60% over baseline approaches.