BRICKS-WM: Building Reusability via Interface Composition Kinetics for Structured World Models

2026-06-15

AI Summary

This work addresses the limited reusability of world models in model-based reinforcement learning, where conventional monolithic latent dynamics models tightly couple agent and environment dynamics, hindering transfer across agents. To overcome this, the authors propose a modular and structured world model framework that decomposes the latent state space and introduces learnable interface mechanisms to disentangle global dynamics into independent background and agent-specific modules. This approach achieves, for the first time, a functional separation of background and agent dynamics at the latent level, enabling the background module to be frozen and reused across different agents. Experimental results demonstrate that the proposed framework matches the performance of strong baselines when trained from scratch while substantially improving the reusability of the learned world model.

Abstract

Model-based Reinforcement Learning (MBRL) has achieved remarkable success in continuous control by leveraging latent world models. However, prevailing approaches typically rely on monolithic latent dynamics, entangling environment dynamics into a coupled process. This coupling severely limits reusability: altering the agent necessitates retraining the entire world from scratch, even if the environment remains constant. To address this, we introduce BRICKS-WM (Building Reusability via Interface Composition Kinetics for Structured World Models), a framework for the modular assembly of structured world models. Driven by the insight that the physical world is composed of independent entities, we posit that global dynamics can be modeled as a composition of distinct dynamical modules interacting via latent interfaces. As a minimal instantiation, we factorize the latent state space into an actuated Agent module and an external Background module, bridged by a learned latent interface. Unlike prior object-centric methods that prioritize visual segmentation, BRICKS-WM enforces a functional separation in transition dynamics, ensuring that background dynamics remains agnostic to the agent's dynamics. Empirically, BRICKS-WM achieves control performance comparable to strong monolithic baselines when trained from scratch, and enables the reuse of frozen background dynamics across agents.

Problem

• Model-based Reinforcement Learning
• World Models
• Reusability
• Modular Dynamics
• Latent Dynamics

Innovation

• modular world models
• dynamics disentanglement
• latent interface
• model-based reinforcement learning
• reusability