Saved in:
Bibliographic Details
Main Authors: Huang, Yihong, Chu, Chen
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.12683
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908369395122176
author Huang, Yihong
Chu, Chen
author_facet Huang, Yihong
Chu, Chen
contents Key feature fields need bigger embedding dimensionality, others need smaller. This demands automated dimension allocation. Existing approaches, such as pruning or Neural Architecture Search (NAS), require training a memory-intensive SuperNet that enumerates all possible dimension combinations, which is infeasible for large feature spaces. We propose DimGrow, a lightweight approach that eliminates the SuperNet requirement. Starting training model from one dimension per feature field, DimGrow can progressively expand/shrink dimensions via importance scoring. Dimensions grow only when their importance consistently exceed a threshold, ensuring memory efficiency. Experiments on three recommendation datasets verify the effectiveness of DimGrow while it reduces training memory compared to SuperNet-based methods.
format Preprint
id arxiv_https___arxiv_org_abs_2505_12683
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DimGrow: Memory-Efficient Field-level Embedding Dimension Search
Huang, Yihong
Chu, Chen
Machine Learning
Key feature fields need bigger embedding dimensionality, others need smaller. This demands automated dimension allocation. Existing approaches, such as pruning or Neural Architecture Search (NAS), require training a memory-intensive SuperNet that enumerates all possible dimension combinations, which is infeasible for large feature spaces. We propose DimGrow, a lightweight approach that eliminates the SuperNet requirement. Starting training model from one dimension per feature field, DimGrow can progressively expand/shrink dimensions via importance scoring. Dimensions grow only when their importance consistently exceed a threshold, ensuring memory efficiency. Experiments on three recommendation datasets verify the effectiveness of DimGrow while it reduces training memory compared to SuperNet-based methods.
title DimGrow: Memory-Efficient Field-level Embedding Dimension Search
topic Machine Learning
url https://arxiv.org/abs/2505.12683