Neural Production Systems: Learning Rule-Governed Visual Dynamics

TL;DR

This paper introduces a neural production system architecture inspired by cognitive science, which models object interactions in visual environments more effectively than GNNs, enabling better future prediction and scalability.

Contribution

The paper proposes a novel neural production system that factorizes entity-specific and rule-based knowledge, outperforming GNNs in modeling visual dynamics.

Findings

Achieves robust future-state prediction in complex visual environments.

Outperforms state-of-the-art GNN-based methods.

Enables extrapolation from simple to complex environments.

Abstract

Visual environments are structured, consisting of distinct objects or entities. These entities have properties -- both visible and latent -- that determine the manner in which they interact with one another. To partition images into entities, deep-learning researchers have proposed structural inductive biases such as slot-based architectures. To model interactions among entities, equivariant graph neural nets (GNNs) are used, but these are not particularly well suited to the task for two reasons. First, GNNs do not predispose interactions to be sparse, as relationships among independent entities are likely to be. Second, GNNs do not factorize knowledge about interactions in an entity-conditional manner. As an alternative, we take inspiration from cognitive science and resurrect a classic approach, production systems, which consist of a set of rule templates that are applied by binding placeholder variables in the rules to specific entities. Rules are scored on their match to entities, and the best fitting rules are applied to update entity properties. In a series of experiments, we demonstrate that this architecture achieves a flexible, dynamic flow of control and serves to factorize entity-specific and rule-based information. This disentangling of knowledge achieves robust future-state prediction in rich visual environments, outperforming state-of-the-art methods using GNNs, and allows for the extrapolation from simple (few object) environments to more complex environments.