Noether Networks: Meta-Learning Useful Conserved Quantities
arXiv:2112.03321
Abstract
Progress in machine learning (ML) stems from a combination of data availability, computational resources, and an appropriate encoding of inductive biases. Useful biases often exploit symmetries in the prediction problem, such as convolutional networks relying on translation equivariance. Automatically discovering these useful symmetries holds the potential to greatly improve the performance of ML systems, but still remains a challenge. In this work, we focus on sequential prediction problems and take inspiration from Noether's theorem to reduce the problem of finding inductive biases to meta-learning useful conserved quantities. We propose Noether Networks: a new type of architecture where a meta-learned conservation loss is optimized inside the prediction function. We show, theoretically and experimentally, that Noether Networks improve prediction quality, providing a general framework for discovering inductive biases in sequential problems.
arXiv commentAccepted to NeurIPS '21. The first two authors contributed equally
References in corpus (8)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Learning to Simulate Complex Physics with Graph Networks
- Stochastic Video Generation with a Learned Prior
- One-Shot Imitation from Observing Humans via Domain-Adaptive Meta-Learning
- Symbolic Pregression: Discovering Physical Laws from Distorted Video
- Port-Hamiltonian Neural Networks for Learning Explicit Time-Dependent Dynamical Systems
- Neural Relational Inference with Fast Modular Meta-learning