Exploring Covariate and Concept Shift for Detection and Calibration of Out-of-Distribution Data

Abstract

Moving beyond testing on in-distribution data works on Out-of-Distribution (OOD) detection have recently increased in popularity. A recent attempt to categorize OOD data introduces the concept of near and far OOD detection. Specifically, prior works define characteristics of OOD data in terms of detection difficulty. We propose to characterize the spectrum of OOD data using two types of distribution shifts: covariate shift and concept shift, where covariate shift corresponds to change in style, e.g., noise, and concept shift indicates a change in semantics. This characterization reveals that sensitivity to each type of shift is important to the detection and confidence calibration of OOD data. Consequently, we investigate score functions that capture sensitivity to each type of dataset shift and methods that improve them. To this end, we theoretically derive two score functions for OOD detection, the covariate shift score and concept shift score, based on the decomposition of KL-divergence for both scores, and propose a geometrically-inspired method (Geometric ODIN) to improve OOD detection under both shifts with only in-distribution data. Additionally, the proposed method naturally leads to an expressive post-hoc calibration function which yields state-of-the-art calibration performance on both in-distribution and out-of-distribution data. We are the first to propose a method that works well across both OOD detection and calibration and under different types of shifts. View project page at https://sites.google.com/view/geometric-decomposition.

arXiv commentA short version of the paper is accepted to NeurIPS DistShift Workshop 2021

References in corpus (24)

• Spectral Normalization for Generative Adversarial Networks
• End to End Learning for Self-Driving Cars
• Wide Residual Networks
• On Calibration of Modern Neural Networks
• SGDR: Stochastic Gradient Descent with Warm Restarts
• LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop
• A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks
• Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift
• Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles
• Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networks
• Deep Anomaly Detection with Outlier Exposure
• Uncertainty Estimation Using a Single Deep Deterministic Neural Network
• CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances
• TurkerGaze: Crowdsourcing Saliency with Webcam based Eye Tracking
• Decomposition of Uncertainty in Bayesian Deep Learning for Efficient and Risk-sensitive Learning
• Exploring the Limits of Out-of-Distribution Detection
• Simple and Principled Uncertainty Estimation with Deterministic Deep Learning via Distance Awareness
• Verified Uncertainty Calibration
• Does Your Dermatology Classifier Know What It Doesn't Know? Detecting the Long-Tail of Unseen Conditions
• Contrastive Training for Improved Out-of-Distribution Detection
• Unlabelled Data Improves Bayesian Uncertainty Calibration under Covariate Shift
• Hyperparameter-Free Out-of-Distribution Detection Using Softmax of Scaled Cosine Similarity
• A Geometric Perspective towards Neural Calibration via Sensitivity Decomposition
• Memory-Efficient Semi-Supervised Continual Learning: The World is its Own Replay Buffer