P-KDGAN: Progressive Knowledge Distillation with GANs for One-class Novelty Detection

TL;DR

This paper introduces P-KDGAN, a progressive knowledge distillation method for GAN-based one-class novelty detection, achieving high compression ratios and improved detection performance on standard datasets.

Contribution

It proposes a novel progressive knowledge distillation framework connecting two GANs for efficient and accurate one-class novelty detection.

Findings

Improves student GAN performance by up to 2.44% on CIFAR-10.

Achieves high compression ratios, e.g., 700:1, with minimal performance loss.

Demonstrates state-of-the-art results on multiple datasets.

Abstract

One-class novelty detection is to identify anomalous instances that do not conform to the expected normal instances. In this paper, the Generative Adversarial Networks (GANs) based on encoder-decoder-encoder pipeline are used for detection and achieve state-of-the-art performance. However, deep neural networks are too over-parameterized to deploy on resource-limited devices. Therefore, Progressive Knowledge Distillation with GANs (PKDGAN) is proposed to learn compact and fast novelty detection networks. The P-KDGAN is a novel attempt to connect two standard GANs by the designed distillation loss for transferring knowledge from the teacher to the student. The progressive learning of knowledge distillation is a two-step approach that continuously improves the performance of the student GAN and achieves better performance than single step methods. In the first step, the student GAN learns the basic knowledge totally from the teacher via guiding of the pretrained teacher GAN with fixed weights. In the second step, joint fine-training is adopted for the knowledgeable teacher and student GANs to further improve the performance and stability. The experimental results on CIFAR-10, MNIST, and FMNIST show that our method improves the performance of the student GAN by 2.44%, 1.77%, and 1.73% when compressing the computation at ratios of 24.45:1, 311.11:1, and 700:1, respectively.