Efficient Low-Dimensional Compression of Overparameterized Models

被引:0
|
作者
Kwon, Soo Min [1 ]
Zhang, Zekai [2 ]
Song, Dogyoon [1 ]
Balzano, Laura [1 ]
Qu, Qing [1 ]
机构
[1] Univ Michigan, Ann Arbor, MI 48109 USA
[2] Tsinghua Univ, Beijing, Peoples R China
来源
INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 238 | 2024年 / 238卷
关键词
RANK;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In this work, we present a novel approach for compressing overparameterized models, developed through studying their learning dynamics. We observe that for many deep models, updates to the weight matrices occur within a low-dimensional invariant subspace. For deep linear models, we demonstrate that their principal components are fitted incrementally within a small subspace, and use these insights to propose a compression algorithm for deep linear networks that involve decreasing the width of their intermediate layers. We empirically evaluate the effectiveness of our compression technique on matrix recovery problems. Remarkably, by using an initialization that exploits the structure of the problem, we observe that our compressed network converges faster than the original network, consistently yielding smaller recovery errors. We substantiate this observation by developing a theory focused on deep matrix factorization. Finally, we empirically demonstrate how our compressed model has the potential to improve the utility of deep nonlinear models. Overall, our algorithm improves the training efficiency by more than 2x, without compromising generalization.
引用
收藏
页数:26
相关论文
empty
未找到相关数据