Acquiring Human-Like Mechanics Intuition from Scarce Observations via Deep Reinforcement Learning

TL;DR

This paper introduces a deep reinforcement learning framework that enables agents to develop human-like mechanics intuition from very few observations, generalizing across wide parameter ranges.

Contribution

It presents a novel episodic switching training method that fosters robust physical intuition and provides a theoretical explanation for its generalization capabilities.

Findings

Agents acquire mechanics intuition with only 2-3 observations

The learned value function enforces Bellman consistency across parameters

The approach generalizes well beyond training data

Abstract

Humans can infer accurate mechanical outcomes from only a few observations, a capability known as mechanics intuition. The mechanisms behind such data-efficient learning remain unclear. Here, we propose a reinforcement learning framework in which an agent encodes continuous physical observation parameters into its state and is trained via episodic switching across closely related observations. With merely two or three observations, the agent acquires robust mechanics intuition that generalizes accurately over wide parameter ranges, substantially beyond the training data, as demonstrated on the brachistochrone and a large-deformation elastic plate. We explain this generalization through a unified theoretical view: it emerges when the learned value function enforces Bellman consistency across neighboring task parameters, rendering the Bellman residual stationary with respect to physical variations. This induces a smooth policy that captures a low-dimensional solution manifold underlying the continuum of tasks. Our work establishes episodic switching as a principled route to artificial mechanics intuition and offers a theoretical link to similar generalization abilities in biological learners.