CoPhy: Counterfactual Learning of Physical Dynamics

TL;DR

This paper introduces CoPhy, a benchmark and model for counterfactual learning of physical dynamics from visual input, enabling prediction of alternative outcomes after interventions in mechanical systems.

Contribution

It presents a new benchmark and a novel model for counterfactual physical reasoning from visual data, outperforming existing methods.

Findings

Model achieves superhuman prediction accuracy.

Learning latent physical properties improves outcome predictions.

Counterfactual reasoning enhances understanding of physical systems.

Abstract

Understanding causes and effects in mechanical systems is an essential component of reasoning in the physical world. This work poses a new problem of counterfactual learning of object mechanics from visual input. We develop the CoPhy benchmark to assess the capacity of the state-of-the-art models for causal physical reasoning in a synthetic 3D environment and propose a model for learning the physical dynamics in a counterfactual setting. Having observed a mechanical experiment that involves, for example, a falling tower of blocks, a set of bouncing balls or colliding objects, we learn to predict how its outcome is affected by an arbitrary intervention on its initial conditions, such as displacing one of the objects in the scene. The alternative future is predicted given the altered past and a latent representation of the confounders learned by the model in an end-to-end fashion with no supervision. We compare against feedforward video prediction baselines and show how observing alternative experiences allows the network to capture latent physical properties of the environment, which results in significantly more accurate predictions at the level of super human performance.