Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2410.03989 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914965473984512 |
|---|---|
| author | Huang, Haozhe Cheng, Leo Kaixuan Chen, Kaiwen Aspuru-Guzik, Alán |
| author_facet | Huang, Haozhe Cheng, Leo Kaixuan Chen, Kaiwen Aspuru-Guzik, Alán |
| contents | In machine learning datasets with symmetries, the paradigm for backward compatibility with symmetry-breaking has been to relax equivariant architectural constraints, engineering extra weights to differentiate symmetries of interest. However, this process becomes increasingly over-engineered as models are geared towards specific symmetries/asymmetries hardwired of a particular set of equivariant basis functions. In this work, we introduce symmetry-cloning, a method for inducing equivariance in machine learning models. We show that general machine learning architectures (i.e., MLPs) can learn symmetries directly as a supervised learning task from group equivariant architectures and retain/break the learned symmetry for downstream tasks. This simple formulation enables machine learning models with group-agnostic architectures to capture the inductive bias of group-equivariant architectures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_03989 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Symmetry From Scratch: Group Equivariance as a Supervised Learning Task Huang, Haozhe Cheng, Leo Kaixuan Chen, Kaiwen Aspuru-Guzik, Alán Machine Learning In machine learning datasets with symmetries, the paradigm for backward compatibility with symmetry-breaking has been to relax equivariant architectural constraints, engineering extra weights to differentiate symmetries of interest. However, this process becomes increasingly over-engineered as models are geared towards specific symmetries/asymmetries hardwired of a particular set of equivariant basis functions. In this work, we introduce symmetry-cloning, a method for inducing equivariance in machine learning models. We show that general machine learning architectures (i.e., MLPs) can learn symmetries directly as a supervised learning task from group equivariant architectures and retain/break the learned symmetry for downstream tasks. This simple formulation enables machine learning models with group-agnostic architectures to capture the inductive bias of group-equivariant architectures. |
| title | Symmetry From Scratch: Group Equivariance as a Supervised Learning Task |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2410.03989 |