Towards Oracle Knowledge Distillation with Neural Architecture Search

TL;DR

This paper introduces a neural architecture search-based framework for knowledge distillation that effectively reduces the capacity gap between teacher and student models, leading to improved performance and efficiency.

Contribution

It proposes a novel neural architecture search method combined with oracle knowledge distillation to enhance student model learning from ensemble teachers.

Findings

Searched models outperform their teachers in accuracy.

The approach reduces the capacity gap between teacher and student.

Effective in both accuracy and memory size improvements.

Abstract

We present a novel framework of knowledge distillation that is capable of learning powerful and efficient student models from ensemble teacher networks. Our approach addresses the inherent model capacity issue between teacher and student and aims to maximize benefit from teacher models during distillation by reducing their capacity gap. Specifically, we employ a neural architecture search technique to augment useful structures and operations, where the searched network is appropriate for knowledge distillation towards student models and free from sacrificing its performance by fixing the network capacity. We also introduce an oracle knowledge distillation loss to facilitate model search and distillation using an ensemble-based teacher model, where a student network is learned to imitate oracle performance of the teacher. We perform extensive experiments on the image classification datasets---CIFAR-100 and TinyImageNet---using various networks. We also show that searching for a new student model is effective in both accuracy and memory size and that the searched models often outperform their teacher models thanks to neural architecture search with oracle knowledge distillation.