Equivariant modeling framework speeds scientific predictions with fewer parameters

Equivariant Modeling with Polynomial Features via Minimal Homogeneous Generators

Machine Learning

Summary

Predicting how things change when you move or rotate them is tricky, but important in science and engineering. The authors developed a general way to build models that keep track of these changes correctly using special polynomial features. Their approach finds the simplest building blocks to represent these changes and learns from data efficiently. They also created software that automates this process, making it faster than existing methods. This helps make accurate predictions with smaller, easier-to-manage models.

What this means in practice

Authors

Michael Liaofan Liu, Joe Kileel

Abstract

We present a general framework for modeling and learning equivariant functions between finite dimensional semisimple real group representations, unifying constructions developed across many application areas. Our model is a linear combination of equivariant polynomial features whose coefficients are arbitrary learnable functions of invariant polynomial features. We take as features minimal homogeneous generators for the invariant algebra and the covariant module over the invariant algebra, up to prescribed degrees. We provide a Python package that automatically computes these features and fits the resulting models to training data. The algorithm for feature computation requires explicit isotypic decompositions of the input and output representations, along with explicit Clebsch-Gordan decompositions of the tensor product of irreducible representations, and achieves speedups over alternative methods in computational algebra. Numerical experiments in scientific modeling demonstrate the full learning pipeline and yield accurate predictors with few trainable parameters, using the model both for end-to-end prediction and within layers of larger architectures.