Causality Meets Locality: Provably Generalizable and Scalable Policy Learning for Networked Systems

TL;DR

This paper introduces GSAC, a framework combining causal representation learning with meta actor-critic methods to enable scalable, generalizable policy learning in large networked systems with environment shifts.

Contribution

The paper proposes a novel GSAC framework that learns sparse local causal masks and compact domain factors, providing provable guarantees and efficient adaptation for networked systems.

Findings

GSAC achieves rapid adaptation to new domains with few trajectories.

The method outperforms learning-from-scratch and baseline approaches.

Finite-sample guarantees are established for causal recovery and policy convergence.

Abstract

Large-scale networked systems, such as traffic, power, and wireless grids, challenge reinforcement-learning agents with both scale and environment shifts. To address these challenges, we propose GSAC (Generalizable and Scalable Actor-Critic), a framework that couples causal representation learning with meta actor-critic learning to achieve both scalability and domain generalization. Each agent first learns a sparse local causal mask that provably identifies the minimal neighborhood variables influencing its dynamics, yielding exponentially tight approximately compact representations (ACRs) of state and domain factors. These ACRs bound the error of truncating value functions to $\kappa$-hop neighborhoods, enabling efficient learning on graphs. A meta actor-critic then trains a shared policy across multiple source domains while conditioning on the compact domain factors; at test time, a few trajectories suffice to estimate the new domain factor and deploy the adapted policy. We establish finite-sample guarantees on causal recovery, actor-critic convergence, and adaptation gap, and show that GSAC adapts rapidly and significantly outperforms learning-from-scratch and conventional adaptation baselines.