Newest papers

From Observation to Intervention: Mastering Bus Holding Control from Historical Uncontrolled Trajectories with Offline Reinforcement Learning

Minyu Shen, Xiaowen Bi, Weihua Gu, Zhen Xu, Yuezhong Wu, Feng Xiao, Lu Li · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.14632 · cited by 0 works (OpenAlex) · open-access PDF

Vietnamese

Hoạt động xe buýt mục tiêu thậm chí là dữ liệu không ổn định---các xe buýt ban đầu được triển khai ở các tuyến đường ngang sẽ không thể tránh khỏi kết thúc bằng cách tụt lại. Vấn đề này làm trầm trọng thêm thời gian chờ hành khách, tạo ra sự mất cân bằng tải giữa các xe buýt và làm suy yếu độ tin cậy của dịch vụ. Học tập tăng cường (RL) đã được chứng minh là rất hiệu quả trong việc giảm bớt việc tụt xe buýt và vượt trội hơn các phương pháp liên kết dựa trên mô hình truyền thống. Tuy nhiên, các phương pháp RL hiện có đòi hỏi các đại lý liên lạc liên tục với môi trường, có khả năng phá vỡ hệ thống xe buýt trước khi các chính sách hiệu quả được học. Bài báo này đề xuất một phương pháp Multi-agent Actor-Critic (Offline-GMAC) dựa trên

Machine translation (Meta M2M100); the English original follows.

Abstract

Bus operations are inherently unstable---buses initially dispatched at equal headways will inevitably end up bunching up. This issue exacerbates passenger waiting times, creates load imbalances between buses, and undermines service reliability. Reinforcement learning (RL) has proven highly effective for mitigating bus bunching and outperforms traditional model-based approaches. However, existing RL approaches require agents to continuously interact with the environment, which potentially disrupts the bus system before effective policies are learned. This paper proposes an Offline Graph-based Multi-agent Actor-Critic (Offline-GMAC) method that trains bus holding policies using only readily available logged GPS data collected during normal uncontrolled operations, thereby eliminating the need for disruptive online interaction or exploration. The method employs general graph neural networks for multi-agent credit assignment. A key innovation of our method is a data augmentation scheme that synthesizes unperformed holding actions from historical uncontrolled bus trajectories. The scheme exploits the equivalence between stochastic link travel time delays and deliberate holding interventions. Under mild assumptions, we prove that the 1-Wasserstein distance between synthesized and target state transition distributions is bounded above by a term that scales linearly with the standard deviation of link travel times plus a tunable synthesis-dependent term. Evaluations using real-world bus data demonstrate that the proposed offline method achieves performance comparable to, and in some cases, even better than its online counterpart.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Reinforcement learning, Train, Exploit, Key (lock), Scheme (mathematics)

Is this paper real? This page was built from OpenAlex and arXiv records, not generated. Software can verify any reference in one call: GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.14632, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txt

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange