Learning to Represent Mechanics via Long-term Extrapolation and Interpolation
This work addresses the challenge of automating physical modeling in AI for complex real-world scenarios, though it is incremental in improving upon existing neural network approaches.
The paper tackled the problem of enabling neural networks to learn physical states from visual data alone for long-term motion extrapolation, achieving competitive results with methods that require internal physics models.
While the basic laws of Newtonian mechanics are well understood, explaining a physical scenario still requires manually modeling the problem with suitable equations and associated parameters. In order to adopt such models for artificial intelligence, researchers have handcrafted the relevant states, and then used neural networks to learn the state transitions using simulation runs as training data. Unfortunately, such approaches can be unsuitable for modeling complex real-world scenarios, where manually authoring relevant state spaces tend to be challenging. In this work, we investigate if neural networks can implicitly learn physical states of real-world mechanical processes only based on visual data, and thus enable long-term physical extrapolation. We develop a recurrent neural network architecture for this task and also characterize resultant uncertainties in the form of evolving variance estimates. We evaluate our setup to extrapolate motion of a rolling ball on bowl of varying shape and orientation using only images as input, and report competitive results with approaches that assume access to internal physics models and parameters.