Deep Residual Flow for Novelty Detection

01/15/2020
by   Ev Zisselman, et al.
3

The effective application of neural networks in the real-world relies on proficiently detecting out-of-distribution examples. Contemporary methods seek to model the distribution of feature activations in the training data for adequately distinguishing abnormalities, and the state-of-the-art method uses Gaussian distribution models. In this work, we present a novel approach that improves upon the state-of-the-art by leveraging an expressive density model based on normalizing flows. We introduce the residual flow, a novel flow architecture that learns the residual distribution from a base Gaussian distribution. Our model is general, and can be applied to any data that is approximately Gaussian. For novelty detection in image datasets, our approach provides a principled improvement over the state-of-the-art. Specifically, we demonstrate the effectiveness of our method in ResNet and DenseNet architectures trained on various image datasets. For example, on a ResNet trained on CIFAR-100 and evaluated on detection of out-of-distribution samples from the ImageNet dataset, holding the true positive rate (TPR) at 95%, we improve the true negative rate (TNR) from 56.7% (current state-of-the-art) to 77.5% (ours).

READ FULL TEXT
research
03/25/2016

Resnet in Resnet: Generalizing Residual Architectures

Residual networks (ResNets) have recently achieved state-of-the-art on c...
research
10/25/2022

Connective Reconstruction-based Novelty Detection

Detection of out-of-distribution samples is one of the critical tasks fo...
research
08/16/2020

GLOD: Gaussian Likelihood Out of Distribution Detector

Discriminative deep neural networks (DNNs) do well at classifying input ...
research
09/09/2023

HAct: Out-of-Distribution Detection with Neural Net Activation Histograms

We propose a simple, efficient, and accurate method for detecting out-of...
research
06/17/2019

Normalizing flows for novelty detection in industrial time series data

Flow-based deep generative models learn data distributions by transformi...
research
05/28/2022

Fake It Till You Make It: Near-Distribution Novelty Detection by Score-Based Generative Models

We aim for image-based novelty detection. Despite considerable progress,...
research
08/23/2022

Semantic Driven Energy based Out-of-Distribution Detection

Detecting Out-of-Distribution (OOD) samples in real world visual applica...

Please sign up or login with your details

Forgot password? Click here to reset