stat.MLMay 4, 2026

Middle-mile logistics through the lens of goal-conditioned reinforcement learning

Authors: Onno EberhardThibaut CuvelierMichal ValkoBruno De Backer

Organizations: Max Planck Institute for Intelligent Systems, Tübingen, Germany · University of Tübingen · Google Research · Google DeepMind

Abstract

Middle-mile logistics describes the problem of routing parcels through a network of hubs linked by trucks with finite capacity. We rephrase this as a multi-object goal-conditioned MDP. Our method combines graph neural networks with model-free RL, extracting small feature graphs from the environment state.

Explore similar work

CardsList