Novelty Detection with GAN

Abstract

The ability of a classifier to recognize unknown inputs is important for many classification-based systems. We discuss the problem of simultaneous classification and novelty detection, i.e. determining whether an input is from the known set of classes and from which specific class, or from an unknown domain and does not belong to any of the known classes. We propose a method based on the Generative Adversarial Networks (GAN) framework. We show that a multi-class discriminator trained with a generator that generates samples from a mixture of nominal and novel data distributions is the optimal novelty detector. We approximate that generator with a mixture generator trained with the Feature Matching loss and empirically show that the proposed method outperforms conventional methods for novelty detection. Our findings demonstrate a simple, yet powerful new application of the GAN framework for the task of novelty detection.

References in corpus (9)

• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Generative Adversarial Text to Image Synthesis
• A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks
• Improved Techniques for Training GANs
• NIPS 2016 Tutorial: Generative Adversarial Networks
• Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks
• Semi-Supervised Learning with Generative Adversarial Networks
• Unsupervised and Semi-supervised Learning with Categorical Generative Adversarial Networks
• Good Semi-supervised Learning that Requires a Bad GAN

Cited by in corpus (15)

• Deep Learning for Anomaly Detection: A Survey
• WAIC, but Why? Generative Ensembles for Robust Anomaly Detection
• Outlier Exposure with Confidence Control for Out-of-Distribution Detection
• Safer Classification by Synthesis
• A Less Biased Evaluation of Out-of-distribution Sample Detectors
• Unsupervised Out-of-Distribution Detection with Batch Normalization
• OMNIA Faster R-CNN: Detection in the wild through dataset merging and soft distillation
• Imbalanced Data Learning by Minority Class Augmentation using Capsule Adversarial Networks
• Out-of-domain Detection for Natural Language Understanding in Dialog Systems
• Oversampling Adversarial Network for Class-Imbalanced Fault Diagnosis
• Enhanced Isotropy Maximization Loss: Seamless and High-Performance Out-of-Distribution Detection Simply Replacing the SoftMax Loss
• AMAD: Adversarial Multiscale Anomaly Detection on High-Dimensional and Time-Evolving Categorical Data
• Unsupervised Class-Incremental Learning Through Confusion
• ${\ell }_p$-Norm Multiple Kernel One-Class Fisher Null-Space
• Image-based OoD-Detector Principles on Graph-based Input Data in Human Action Recognition