Data-Free Knowledge Distillation for Deep Neural Networks

TL;DR

This paper introduces a data-free knowledge distillation method that compresses large neural networks without access to original training data, using only metadata, addressing privacy and data availability issues.

Contribution

The authors propose a novel data-free knowledge distillation approach that leverages metadata to effectively compress neural networks without original data.

Findings

Effective compression of neural networks without original data

Different types of metadata can be used with tradeoffs

Method preserves accuracy comparable to data-dependent approaches

Abstract

Recent advances in model compression have provided procedures for compressing large neural networks to a fraction of their original size while retaining most if not all of their accuracy. However, all of these approaches rely on access to the original training set, which might not always be possible if the network to be compressed was trained on a very large dataset, or on a dataset whose release poses privacy or safety concerns as may be the case for biometrics tasks. We present a method for data-free knowledge distillation, which is able to compress deep neural networks trained on large-scale datasets to a fraction of their size leveraging only some extra metadata to be provided with a pretrained model release. We also explore different kinds of metadata that can be used with our method, and discuss tradeoffs involved in using each of them.