Approximate exploitability: Learning a best response in large games
CoRR(2020)
摘要
Researchers have demonstrated that neural networks are vulnerable to
adversarial examples and subtle environment changes, both of which one can view
as a form of distribution shift. To humans, the resulting errors can look like
blunders, eroding trust in these agents. In prior games research, agent
evaluation often focused on the in-practice game outcomes. While valuable, such
evaluation typically fails to evaluate robustness to worst-case outcomes. Prior
research in computer poker has examined how to assess such worst-case
performance, both exactly and approximately. Unfortunately, exact computation
is infeasible with larger domains, and existing approximations rely on
poker-specific knowledge. We introduce ISMCTS-BR, a scalable search-based deep
reinforcement learning algorithm for learning a best response to an agent,
thereby approximating worst-case performance. We demonstrate the technique in
several two-player zero-sum games against a variety of agents, including
several AlphaZero-based agents.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要