Causal Process Models: Reframing Dynamic Causal Graph Discovery as a Reinforcement Learning Problem

TL;DR

This paper introduces Causal Process Models, a reinforcement learning approach for learning sparse, dynamic causal graphs from visual data, improving interpretability and efficiency in physical world modeling.

Contribution

It presents a novel neural framework that constructs time-varying causal graphs via reinforcement learning, with a structured representation for semantic encoding.

Findings

CPMs outperform dense graph baselines on physical prediction tasks

The model effectively captures long-term dependencies and varying object counts

Structured representations enable meaningful causal encodings

Abstract

Most neural models of causality assume static causal graphs, failing to capture the dynamic and sparse nature of physical interactions where causal relationships emerge and dissolve over time. We introduce the Causal Process Framework and its neural implementation, Causal Process Models (CPMs), for learning sparse, time-varying causal graphs from visual observations. Unlike traditional approaches that maintain dense connectivity, our model explicitly constructs causal edges only when objects actively interact, dramatically improving both interpretability and computational efficiency. We achieve this by casting dynamic interaction-graph construction for world modeling as a multi-agent reinforcement learning problem, where specialized agents sequentially decide which objects are causally connected at each timestep. Our key innovation is a structured representation that factorizes object and force vectors along three learned dimensions (mutability, causal relevance, and control relevance), enabling the automatic discovery of semantically meaningful encodings. We demonstrate that a CPM significantly outperforms dense graph baselines on physical prediction tasks, particularly for longer horizons and varying object counts.