Physical Reasoning Using Dynamics-Aware Models

20 Feb 2021  ·  Eltayeb Ahmed, Anton Bakhtin, Laurens van der Maaten, Rohit Girdhar ·

A common approach to solving physical reasoning tasks is to train a value learner on example tasks. A limitation of such an approach is that it requires learning about object dynamics solely from reward values assigned to the final state of a rollout of the environment. This study aims to address this limitation by augmenting the reward value with self-supervised signals about object dynamics. Specifically, we train the model to characterize the similarity of two environment rollouts, jointly with predicting the outcome of the reasoning task. This similarity can be defined as a distance measure between the trajectory of objects in the two rollouts, or learned directly from pixels using a contrastive formulation. Empirically, we find that this approach leads to substantial performance improvements on the PHYRE benchmark for physical reasoning (Bakhtin et al., 2019), establishing a new state-of-the-art.

PDF Abstract


Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Visual Reasoning PHYRE-1B-Cross Dynamics-Aware DQN AUCCESS 39.9 # 3
Visual Reasoning PHYRE-1B-Within Dynamics-Aware DQN AUCCESS 85.2 # 1


No methods listed for this paper. Add relevant methods here