arxiv: 1802.05747 · v2 · pith:RIJBBO2Vnew · submitted 2018-02-15 · 💻 cs.LG · cs.CV

Systematic Weight Pruning of DNNs using Alternating Direction Method of Multipliers

Tianyun Zhang , Shaokai Ye , Yipeng Zhang , Yanzhi Wang , Makan Fardad This is my paper

classification 💻 cs.LG cs.CV

keywords pruningweightadmmdnnssystematicalternatingdirectionframework

0 comments p. Extension

Add this Pith Number to your LaTeX paper

\usepackage{pith}
\pithnumber{RIJBBO2V}

Prints a linked pith:RIJBBO2V badge after your title and writes the identifier into PDF metadata. Compiles on arXiv with no extra files. Learn more

read the original abstract

We present a systematic weight pruning framework of deep neural networks (DNNs) using the alternating direction method of multipliers (ADMM). We first formulate the weight pruning problem of DNNs as a constrained nonconvex optimization problem, and then adopt the ADMM framework for systematic weight pruning. We show that ADMM is highly suitable for weight pruning due to the computational efficiency it offers. We achieve a much higher compression ratio compared with prior work while maintaining the same test accuracy, together with a faster convergence rate. Our models are released at https://github.com/KaiqiZhang/admm-pruning

This paper has not been read by Pith yet.

Systematic Weight Pruning of DNNs using Alternating Direction Method of Multipliers

discussion (0)