arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.12359cs.AIcs.LG
收藏

AlphaZero中的目标搜索控制,用于有效的策略改进

Targeted Search Control in AlphaZero for Effective Policy Improvement

University of Alberta(阿尔伯塔大学)

Alexandre Trudeau, Michael Bowling

摘要:

AlphaZero是一个自我发挥的强化学习算法,通过策略迭代实现国际象棋,将棋和围棋的超人发挥。为了成为一个有效的策略改进算子,AlphaZero的搜索需要对其搜索树中出现的状态进行准确的值估计。AlphaZero从游戏的初始状态开始训练自玩比赛,并且只对前几步的动作进行采样,限制了它对游戏树更深处状态的探索。我们介绍了Go-Exploit,这是AlphaZero的一种新的搜索控制策略。Go-Exploit从感兴趣的状态档案中采样其自我游戏轨迹的开始状态。从不同的起始状态开始自我游戏轨迹使Go-Exploit能够更有效地探索游戏树,并学习更好地概括的价值函数。产生较短的自我发挥轨迹允许去开拓训练更独立的价值目标,提高价值训练。最后,Go-Exploit固有的探索性减少了探索性行动的需要,使其能够在更具剥削性的政策下进行训练。在Connect Four和9 x9 Go的比赛中,我们表明Go-Exploit比标准AlphaZero具有更高的样本效率,从而在对抗参考对手和正面比赛中表现更好。我们还比较了Go-Exploit和KataGo,这是AlphaZero的一个更有效的重新实现,并证明了Go-Exploit具有更有效的搜索控制策略。此外,当KataGo的其他创新被纳入时,Go-Exploit的样本效率得到了提高。

英文摘要:

AlphaZero is a self-play reinforcement learning algorithm that achieves superhuman play in chess, shogi, and Go via policy iteration. To be an effective policy improvement operator, AlphaZero's search requires accurate value estimates for the states appearing in its search tree. AlphaZero trains upon self-play matches beginning from the initial state of a game and only samples actions over the first few moves, limiting its exploration of states deeper in the game tree. We introduce Go-Exploit, a novel search control strategy for AlphaZero. Go-Exploit samples the start state of its self-play trajectories from an archive of states of interest. Beginning self-play trajectories from varied starting states enables Go-Exploit to more effectively explore the game tree and to learn a value function that generalizes better. Producing shorter self-play trajectories allows Go-Exploit to train upon more independent value targets, improving value training. Finally, the exploration inherent in Go-Exploit reduces its need for exploratory actions, enabling it to train under more exploitative policies. In the games of Connect Four and 9x9 Go, we show that Go-Exploit learns with a greater sample efficiency than standard AlphaZero, resulting in stronger performance against reference opponents and in head-to-head play. We also compare Go-Exploit to KataGo, a more sample efficient reimplementation of AlphaZero, and demonstrate that Go-Exploit has a more effective search control strategy. Furthermore, Go-Exploit's sample efficiency improves when KataGo's other innovations are incorporated.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑