Model Compression Using Optimal Transport

Model compression methods are important to allow for easier deployment of deep learning models in compute, memory and energy-constrained environments such as mobile phones. Knowledge distillation is a class of model compression algorithms where knowledge from a large teacher network is transferred t...

Full description

Saved in:
Bibliographic Details
Published in:Proceedings / IEEE Workshop on Applications of Computer Vision pp. 3645 - 3654
Main Authors: Lohit, Suhas, Jones, Michael
Format: Conference Proceeding
Language:English
Published: IEEE 01.01.2022
Subjects:
Computational modeling
Computer vision
Deep learning
Deep Learning -> Efficient Training and Inference Methods for Networks Deep Learning; Object Detection/Recognition/Categorization; Statistical Methods; Learning and Optimization
Image coding
Knowledge engineering
Mobile handsets
Training
ISSN:2642-9381
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Model compression methods are important to allow for easier deployment of deep learning models in compute, memory and energy-constrained environments such as mobile phones. Knowledge distillation is a class of model compression algorithms where knowledge from a large teacher network is transferred to a smaller student network thereby improving the student's performance. In this paper, we show how optimal transport-based loss functions can be used for training a student network which encourages learning student network parameters that help bring the distribution of student features closer to that of the teacher features. We present image classification results on CIFAR-100, SVHN and ImageNet and show that the proposed optimal transport loss functions perform comparably to or better than other loss functions.
ISSN:2642-9381
DOI:10.1109/WACV51458.2022.00370