Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2410.01405 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909637843877888 |
|---|---|
| author | Xu, Kevin Sato, Issei |
| author_facet | Xu, Kevin Sato, Issei |
| contents | Looped Transformers provide advantages in parameter efficiency, computational capabilities, and generalization for reasoning tasks. However, their expressive power regarding function approximation remains underexplored. In this paper, we establish the approximation rate of Looped Transformers by defining the modulus of continuity for sequence-to-sequence functions. This reveals a limitation specific to the looped architecture. That is, the analysis prompts the incorporation of scaling parameters for each loop, conditioned on timestep encoding. Experiments validate the theoretical results, showing that increasing the number of loops enhances performance, with further gains achieved through the timestep encoding. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_01405 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | On Expressive Power of Looped Transformers: Theoretical Analysis and Enhancement via Timestep Encoding Xu, Kevin Sato, Issei Machine Learning Looped Transformers provide advantages in parameter efficiency, computational capabilities, and generalization for reasoning tasks. However, their expressive power regarding function approximation remains underexplored. In this paper, we establish the approximation rate of Looped Transformers by defining the modulus of continuity for sequence-to-sequence functions. This reveals a limitation specific to the looped architecture. That is, the analysis prompts the incorporation of scaling parameters for each loop, conditioned on timestep encoding. Experiments validate the theoretical results, showing that increasing the number of loops enhances performance, with further gains achieved through the timestep encoding. |
| title | On Expressive Power of Looped Transformers: Theoretical Analysis and Enhancement via Timestep Encoding |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2410.01405 |