Relational inductive bias for physical construction in humans and machines

TL;DR

This paper investigates the importance of relational inductive bias in reasoning about object relations, demonstrating that structured representations improve performance in a tower-building task for both humans and machines.

Contribution

It introduces a deep reinforcement learning agent with relation-centric representations that outperforms humans and naive models in a block-stacking task, highlighting the value of relational inductive bias.

Findings

Structured representations enable better task performance.

Relational inductive bias improves reasoning in complex tasks.

The agent outperforms humans in the block-stacking task.

Abstract

While current deep learning systems excel at tasks such as object classification, language processing, and gameplay, few can construct or modify a complex system such as a tower of blocks. We hypothesize that what these systems lack is a "relational inductive bias": a capacity for reasoning about inter-object relations and making choices over a structured description of a scene. To test this hypothesis, we focus on a task that involves gluing pairs of blocks together to stabilize a tower, and quantify how well humans perform. We then introduce a deep reinforcement learning agent which uses object- and relation-centric scene and policy representations and apply it to the task. Our results show that these structured representations allow the agent to outperform both humans and more naive approaches, suggesting that relational inductive bias is an important component in solving structured reasoning problems and for building more intelligent, flexible machines.