Learning Canonical Transformations
arXiv:2011.08822
Abstract
Humans understand a set of canonical geometric transformations (such as translation and rotation) that support generalization by being untethered to any specific object. We explore inductive biases that help a neural network model learn these transformations in pixel space in a way that can generalize out-of-domain. Specifically, we find that high training set diversity is sufficient for the extrapolation of translation to unseen shapes and scales, and that an iterative training scheme achieves significant extrapolation of rotation in time.
arXiv commentNeurIPS 2020 Workshop on BabyMind
References in corpus (6)
- Group Equivariant Convolutional Networks
- Measuring abstract reasoning in neural networks
- Lagrangian Neural Networks
- Environmental drivers of systematicity and generalization in a situated agent
- Learning Through Time in the Thalamocortical Loops
- IterGANs: Iterative GANs to Learn and Control 3D Object Transformation