# Adaptive Value Decomposition: Coordinating a Varying Number of Agents in Urban Systems Source: [Adaptive Value Decomposition: Coordinating a Varying Number of Agents in Urban Systems]() Collection index: [llms.txt]() Data updated: 2026-02-18T05:00:04.391Z Authors: Yexin Li, Jinjin Guo, Haoyu Zhang, Yuhan Zhao, Yiwen Sun, Zihao Jiao Published: 2026-02-10T03:41:14.000Z ## Abstract Multi-agent reinforcement learning (MARL) provides a promising paradigm for coordinating multi-agent systems (MAS). However, most existing methods rely on restrictive assumptions, such as a fixed number of agents and fully synchronous action execution. These assumptions are often violated in urban systems, where the number of active agents varies over time, and actions may have heterogeneous durations, resulting in a semi-MARL setting. Moreover, while sharing policy parameters among agents is commonly adopted to improve learning efficiency, it can lead to highly homogeneous actions when a subset of agents make decisions concurrently under similar observations, potentially degrading coordination quality. To address these challenges, we propose Adaptive Value Decomposition (AVD), a cooperative MARL framework that adapts to a dynamically changing agent population. AVD further incorporates a lightweight mechanism to mitigate action homogenization induced by shared policies, thereby encouraging behavioral diversity and maintaining effective cooperation among agents. In addition, we design a training-execution strategy tailored to the semi-MARL setting that accommodates asynchronous decision-making when some agents act at different times. Experiments on real-world bike-sharing redistribution tasks in two major cities, London and Washington, D.C., demonstrate that AVD outperforms state-of-the-art baselines, confirming its effectiveness and generalizability. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Bike-sharing redistribution on London 20% initial inventory ratio (test)](): Bike-sharing redistribution; London 20% initial inventory ratio (test); Region 1 Count: 1385. - [Bike-sharing redistribution on London 30% initial inventory ratio (test)](): Bike-sharing redistribution; London 30% initial inventory ratio (test); Region 1 Count: 1639. - [Bike Redistribution on Washington D.C. (Region 2)](): Bike Redistribution; Washington D.C. (Region 2); Count @ 5% Threshold: 272. ## Original sources and implementation [Original paper]() [PDF]()