Robust Out-of-Distribution Detection on Deep Probabilistic Generative Models
arXiv:2106.07903
Abstract
Out-of-distribution (OOD) detection is an important task in machine learning systems for ensuring their reliability and safety. Deep probabilistic generative models facilitate OOD detection by estimating the likelihood of a data sample. However, such models frequently assign a suspiciously high likelihood to a specific outlier. Several recent works have addressed this issue by training a neural network with auxiliary outliers, which are generated by perturbing the input data. In this paper, we discover that these approaches fail for certain OOD datasets. Thus, we suggest a new detection metric that operates without outlier exposure. We observe that our metric is robust to diverse variations of an image compared to the previous outlier-exposing methods. Furthermore, our proposed score requires neither auxiliary models nor additional training. Instead, this paper utilizes the likelihood ratio statistic in a new perspective to extract genuine properties from the given single deep probabilistic generative model. We also apply a novel numerical approximation to enable fast implementation. Finally, we demonstrate comprehensive experiments on various probabilistic generative models and show that our method achieves state-of-the-art performance.
arXiv comment23 pages, 11 figures
References in corpus (15)
- Adam: A Method for Stochastic Optimization
- Auto-Encoding Variational Bayes
- Overcoming catastrophic forgetting in neural networks
- Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
- A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks
- Variational Graph Auto-Encoders
- Deep Anomaly Detection with Outlier Exposure
- Learning Confidence for Out-of-Distribution Detection in Neural Networks
- LSTM-based Encoder-Decoder for Multi-sensor Anomaly Detection
- Fast Approximate Natural Gradient Descent in a Kronecker-factored Eigenbasis
- Detecting Out-of-Distribution Inputs to Deep Generative Models Using Typicality
- SSD: A Unified Framework for Self-Supervised Outlier Detection
- Hierarchical VAEs Know What They Don't Know
- Detecting Out-of-distribution Samples via Variational Auto-encoder with Reliable Uncertainty Estimation
- Bigeminal Priors Variational auto-encoder