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 2305.12751cs.SEcs.AIcs.LG
收藏

基于代理模型的深度强化学习智能体测试

Testing of Deep Reinforcement Learning Agents with Surrogate Models

Università della Svizzera italiana(瑞士意大利语区大学)

Matteo Biagiola, Paolo Tonella

AI总结:

本研究提出基于搜索的DRL智能体测试方法,通过工具Indago训练分类器作为代理模型预测失败概率,引导生成失败环境配置,相比现有技术多发现50%的失败,且失败及对应行为多样性分别提升78%、74%。

摘要:

近年来,深度强化学习(DRL)受到了研究界的广泛关注。随着该技术从游戏领域转向自动驾驶、机器人等实际应用场景,评估DRL智能体的质量变得至关重要。本文提出了一种基于搜索的方法来测试这类智能体。我们的方法在名为Indago的工具中实现,它基于DRL训练过程中产生的失败和非失败(即通过)环境配置训练一个分类器。测试时,该分类器作为DRL智能体在环境中执行的代理模型,预测给定环境配置引发被测DRL智能体失败的程度。失败预测结果作为适应度函数,引导生成过程朝向失败环境配置的方向,同时通过将DRL智能体在环境中的执行推迟到更可能暴露失败的配置上,节省计算时间。实验结果表明,我们的基于搜索的方法发现的DRL智能体失败数量比现有最优技术多50%。此外,这些失败的多样性平均高出78%;相应地,失败配置引发的DRL智能体行为的多样性也高出74%。

英文摘要:

Deep Reinforcement Learning (DRL) has received a lot of attention from the research community in recent years. As the technology moves away from game playing to practical contexts, such as autonomous vehicles and robotics, it is crucial to evaluate the quality of DRL agents. In this paper, we propose a search-based approach to test such agents. Our approach, implemented in a tool called Indago, trains a classifier on failure and non-failure environment (i.e., pass) configurations resulting from the DRL training process. The classifier is used at testing time as a surrogate model for the DRL agent execution in the environment, predicting the extent to which a given environment configuration induces a failure of the DRL agent under test. The failure prediction acts as a fitness function, guiding the generation towards failure environment configurations, while saving computation time by deferring the execution of the DRL agent in the environment to those configurations that are more likely to expose failures. Experimental results show that our search-based approach finds 50% more failures of the DRL agent than state-of-the-art techniques. Moreover, such failures are, on average, 78% more diverse; similarly, the behaviors of the DRL agent induced by failure configurations are 74% more diverse.

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。

↑