arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2203.01075cs.LG
收藏

强化学习基准的可靠验证

Reliable validation of Reinforcement Learning Benchmarks

Leiden University(莱顿大学)

Matthias Müller-Brockhausen, Aske Plaat, Mike Preuss

摘要:

强化学习(Reinforcement Learning,RL)是游戏AI和整个AI中最具活力的研究领域之一,各种各样的游戏被用作其突出的测试问题。然而,它受制于目前影响大多数算法AI研究的可复制性危机。强化学习中的基准测试可以通过可验证的结果来改进。有许多基准测试环境,其分数用于比较不同的算法,如Atari。尽管如此,评审人员必须相信数字代表真实的价值,因为很难重现精确的训练曲线。我们建议通过提供原始实验数据来验证研究结果来改善这种情况。为此,我们依赖于最小痕迹的概念。这些允许在确定性RL环境中重新模拟动作序列,反过来,使审查人员能够验证,重用和手动检查实验结果,而无需大型计算集群。它还允许验证呈现的奖励图,检查单个事件,并重新使用结果数据(基线),以便在后续论文中进行适当的比较。我们提供与Gym配合使用的即插即用代码,以便我们的措施能够很好地适应现有的RL和可重复性生态系统。我们的方法是免费提供的,易于使用,并增加了最小的开销,因为最小的跟踪允许高达$\approx 10^4:1$(94 GB到8 MB的Atari Pong)的数据压缩比相比,在离线RL数据集中使用的常规MDP跟踪。本文介绍了各种游戏的概念验证结果。

英文摘要:

Reinforcement Learning (RL) is one of the most dynamic research areas in Game AI and AI as a whole, and a wide variety of games are used as its prominent test problems. However, it is subject to the replicability crisis that currently affects most algorithmic AI research. Benchmarking in Reinforcement Learning could be improved through verifiable results. There are numerous benchmark environments whose scores are used to compare different algorithms, such as Atari. Nevertheless, reviewers must trust that figures represent truthful values, as it is difficult to reproduce an exact training curve. We propose improving this situation by providing access to the original experimental data to validate study results. To that end, we rely on the concept of minimal traces. These allow re-simulation of action sequences in deterministic RL environments and, in turn, enable reviewers to verify, re-use, and manually inspect experimental results without needing large compute clusters. It also permits validation of presented reward graphs, an inspection of individual episodes, and re-use of result data (baselines) for proper comparison in follow-up papers. We offer plug-and-play code that works with Gym so that our measures fit well in the existing RL and reproducibility eco-system. Our approach is freely available, easy to use, and adds minimal overhead, as minimal traces allow a data compression ratio of up to $\approx 10^4:1$ (94GB to 8MB for Atari Pong) compared to a regular MDP trace used in offline RL datasets. The paper presents proof-of-concept results for a variety of games.

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。

↑