Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2305.13404 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913312552255488 |
|---|---|
| author | Zhao, Bo Gower, Robert M. Walters, Robin Yu, Rose |
| author_facet | Zhao, Bo Gower, Robert M. Walters, Robin Yu, Rose |
| contents | In many neural networks, different values of the parameters may result in the same loss value. Parameter space symmetries are loss-invariant transformations that change the model parameters. Teleportation applies such transformations to accelerate optimization. However, the exact mechanism behind this algorithm's success is not well understood. In this paper, we show that teleportation not only speeds up optimization in the short-term, but gives overall faster time to convergence. Additionally, teleporting to minima with different curvatures improves generalization, which suggests a connection between the curvature of the minimum and generalization ability. Finally, we show that integrating teleportation into a wide range of optimization algorithms and optimization-based meta-learning improves convergence. Our results showcase the versatility of teleportation and demonstrate the potential of incorporating symmetry in optimization. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2305_13404 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Improving Convergence and Generalization Using Parameter Symmetries Zhao, Bo Gower, Robert M. Walters, Robin Yu, Rose Machine Learning Optimization and Control In many neural networks, different values of the parameters may result in the same loss value. Parameter space symmetries are loss-invariant transformations that change the model parameters. Teleportation applies such transformations to accelerate optimization. However, the exact mechanism behind this algorithm's success is not well understood. In this paper, we show that teleportation not only speeds up optimization in the short-term, but gives overall faster time to convergence. Additionally, teleporting to minima with different curvatures improves generalization, which suggests a connection between the curvature of the minimum and generalization ability. Finally, we show that integrating teleportation into a wide range of optimization algorithms and optimization-based meta-learning improves convergence. Our results showcase the versatility of teleportation and demonstrate the potential of incorporating symmetry in optimization. |
| title | Improving Convergence and Generalization Using Parameter Symmetries |
| topic | Machine Learning Optimization and Control |
| url | https://arxiv.org/abs/2305.13404 |