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 2212.07313cs.LGcs.MAcs.SYeess.SY
收藏

面向自主按需出行系统的混合多智能体深度强化学习

Hybrid Multi-agent Deep Reinforcement Learning for Autonomous Mobility on Demand Systems

Technical University of Munich(慕尼黑工业大学);Google Research(谷歌研究院);Stanford University(斯坦福大学)

Tobias Enders, James Harrison, Marco Pavone, Maximilian Schiffer

AI总结:

针对自主按需出行系统,提出结合多智能体软演员-评论家与加权二分匹配的前瞻性控制策略,在真实数据上优于现有基准。

摘要:

我们考虑一个利润最大化的自主按需出行系统运营商在序贯决策中主动做出请求分配与拒绝决策的问题。我们将该问题形式化为马尔可夫决策过程,并提出一种多智能体软演员-评论家算法与加权二分匹配的新颖组合,以获得具有前瞻性的控制策略。通过这种方式,我们将运营商原本难以处理的行动空间进行分解,但仍能获得全局协调的决策。基于真实出租车数据的实验表明,我们的方法在性能、稳定性和计算可行性方面均优于现有最先进的基准方法。

英文摘要:

We consider the sequential decision-making problem of making proactive request assignment and rejection decisions for a profit-maximizing operator of an autonomous mobility on demand system. We formalize this problem as a Markov decision process and propose a novel combination of multi-agent Soft Actor-Critic and weighted bipartite matching to obtain an anticipative control policy. Thereby, we factorize the operator's otherwise intractable action space, but still obtain a globally coordinated decision. Experiments based on real-world taxi data show that our method outperforms state of the art benchmarks with respect to performance, stability, and computational tractability.

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。

↑