Addressing Failure Prediction by Learning Model Confidence

Abstract

Assessing reliably the confidence of a deep neural network and predicting its failures is of primary importance for the practical deployment of these models. In this paper, we propose a new target criterion for model confidence, corresponding to the True Class Probability (TCP). We show how using the TCP is more suited than relying on the classic Maximum Class Probability (MCP). We provide in addition theoretical guarantees for TCP in the context of failure prediction. Since the true class is by essence unknown at test time, we propose to learn TCP criterion on the training set, introducing a specific learning scheme adapted to this context. Extensive experiments are conducted for validating the relevance of the proposed approach. We study various network architectures, small and large scale datasets for image classification and semantic segmentation. We show that our approach consistently outperforms several strong methods, from MCP to Bayesian uncertainty, as well as recent approaches specifically designed for failure prediction.

NeurIPS 2019 (accepted)

References in corpus (6)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
• A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks
• ADVENT: Adversarial Entropy Minimization for Domain Adaptation in Semantic Segmentation
• Learning Confidence for Out-of-Distribution Detection in Neural Networks
• Unsupervised Domain Adaptation via Calibrating Uncertainties

Cited by in corpus (11)

• A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
• Run-Time Monitoring of Machine Learning for Robotic Perception: A Survey of Emerging Trends
• Statistical process monitoring of artificial neural networks
• Confidence-Aware Learning for Camouflaged Object Detection
• One-Class Risk Estimation for One-Class Hyperspectral Image Classification
• Collaboration of Experts: Achieving 80% Top-1 Accuracy on ImageNet with 100M FLOPs
• Confidence-Guided Learning Process for Continuous Classification of Time Series
• Learning to Predict Trustworthiness with Steep Slope Loss
• SLURP: Side Learning Uncertainty for Regression Problems
• Thought Flow Nets: From Single Predictions to Trains of Model Thought
• ActiveFreq: Integrating Active Learning and Frequency Domain Analysis for Interactive Segmentation