Saved in:
Bibliographic Details
Main Author: Kozachinskiy, Alexander
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.20195
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910766250065920
author Kozachinskiy, Alexander
author_facet Kozachinskiy, Alexander
contents In this note, we use the VC dimension technique to prove the first lower bound against one-layer softmax transformers with infinite precision. We do so for two tasks: function composition, considered by Peng, Narayanan, and Papadimitriou, and the SUM$_2$ task, considered by Sanford, Hsu, and Telgarsky.
format Preprint
id arxiv_https___arxiv_org_abs_2412_20195
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Lower bounds on transformers with infinite precision
Kozachinskiy, Alexander
Machine Learning
Artificial Intelligence
In this note, we use the VC dimension technique to prove the first lower bound against one-layer softmax transformers with infinite precision. We do so for two tasks: function composition, considered by Peng, Narayanan, and Papadimitriou, and the SUM$_2$ task, considered by Sanford, Hsu, and Telgarsky.
title Lower bounds on transformers with infinite precision
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2412.20195