Addressing Failure Prediction by Learning Model Confidence
Preprint
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.Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- ADVENT: Adversarial Entropy Minimization for Domain Adaptation in Semantic Segmentation2019
- End-to-End Learning of Latent Deformable Part-Based Representations for Object DetectionInternational Journal of Computer Vision, 2018
- Multi-task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics2018
- Efficient Optimization for Rank-Based Loss Functions2018
- Failure Prediction for Autonomous Driving2018
- Focal Loss for Dense Object Detection2017
- MANTRA: Minimum Maximum Latent Structural SVM for Image Classification and Ranking2015
- Deep neural networks are easily fooled: High confidence predictions for unrecognizable images2015
- Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research GroupsIEEE Signal Processing Magazine, 2012
- Neural Network based Intrusion Detection System for critical infrastructures2009