Relational inductive bias for physical construction in humans and machines
This addresses the challenge of structured reasoning in AI for physical construction tasks, offering a novel approach with potential applications in robotics and flexible machine intelligence.
The paper tackles the problem of enabling machines to construct or modify complex physical systems like block towers, which current deep learning systems struggle with, by introducing a deep reinforcement learning agent with relational inductive bias that outperforms both humans and naive approaches on a block-stabilization task.
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.