Addressing Failure Prediction by Learning Model Confidence - Archive ouverte HAL Access content directly
Conference Papers Year :

Addressing Failure Prediction by Learning Model Confidence

(1, 2) , (3, 2) , (2) , (3) , (1)
1
2
3

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.
Fichier principal
Vignette du fichier
8556-addressing-failure-prediction-by-learning-model-confidence.pdf (1.14 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-02469747 , version 1 (06-02-2020)

Identifiers

Cite

Charles Corbière, Nicolas Thome, Avner Bar-Hen, Matthieu Cord, Patrick Pérez. Addressing Failure Prediction by Learning Model Confidence. 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Dec 2019, Vancouver, Canada. pp.2898-2909. ⟨hal-02469747⟩
245 View
161 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More