Trust Issues: Uncertainty Estimation Does Not Enable Reliable OOD Detection On Medical Tabular Data

Abstract

When deploying machine learning models in high-stakes real-world environments such as health care, it is crucial to accurately assess the uncertainty concerning a model's prediction on abnormal inputs. However, there is a scarcity of literature analyzing this problem on medical data, especially on mixed-type tabular data such as Electronic Health Records. We close this gap by presenting a series of tests including a large variety of contemporary uncertainty estimation techniques, in order to determine whether they are able to identify out-of-distribution (OOD) patients. In contrast to previous work, we design tests on realistic and clinically relevant OOD groups, and run experiments on real-world medical data. We find that almost all techniques fail to achieve convincing results, partly disagreeing with earlier findings.

References in corpus (8)

• Adam: A Method for Stochastic Optimization
• A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks
• Weight Uncertainty in Neural Networks
• A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks
• Understanding Measures of Uncertainty for Adversarial Example Detection
• WAIC, but Why? Generative Ensembles for Robust Anomaly Detection
• Likelihood Regret: An Out-of-Distribution Detection Score For Variational Auto-encoder
• Why Normalizing Flows Fail to Detect Out-of-Distribution Data

Cited by in corpus (4)

• A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
• Prior and Posterior Networks: A Survey on Evidential Deep Learning Methods For Uncertainty Estimation
• Deep Anti-Regularized Ensembles provide reliable out-of-distribution uncertainty quantification
• BEDS-Bench: Behavior of EHR-models under Distributional Shift--A Benchmark