A Differentiable Framework for End-to-End Learning of Hybrid Structured Compression

TL;DR

This paper introduces a differentiable framework that unifies filter pruning and low-rank decomposition into an end-to-end trainable system, significantly improving hybrid structured compression performance.

Contribution

It develops a novel differentiable framework with DML-S and DTL-S for integrated filter and rank selection, enabling effective end-to-end hybrid compression.

Findings

Outperforms state-of-the-art structured compression methods

Enables end-to-end gradient-based optimization for hybrid techniques

Provides a versatile approach for structured model compression

Abstract

Filter pruning and low-rank decomposition are two of the foundational techniques for structured compression. Although recent efforts have explored hybrid approaches aiming to integrate the advantages of both techniques, their performance gains have been modest at best. In this study, we develop a \textit{Differentiable Framework~(DF)} that can express filter selection, rank selection, and budget constraint into a single analytical formulation. Within the framework, we introduce DML-S for filter selection, integrating scheduling into existing mask learning techniques. Additionally, we present DTL-S for rank selection, utilizing a singular value thresholding operator. The framework with DML-S and DTL-S offers a hybrid structured compression methodology that facilitates end-to-end learning through gradient-base optimization. Experimental results demonstrate the efficacy of DF, surpassing state-of-the-art structured compression methods. Our work establishes a robust and versatile avenue for advancing structured compression techniques.