Bayesian Metric Learning for Uncertainty Quantification in Image Retrieval

Frederik Rahbæk Warburg; Marco Miani; Silas Brack; Søren Hauberg

NeurIPS 2023poster14 citations

Bayesian Metric Learning for Uncertainty Quantification in Image Retrieval

Frederik Rahbæk Warburg, Marco Miani, Silas Brack, Søren Hauberg

Abstract

We propose a Bayesian encoder for metric learning. Rather than relying on neural amortization as done in prior works, we learn a distribution over the network weights with the Laplace Approximation. We first prove that the contrastive loss is a negative log-likelihood on the spherical space. We propose three methods that ensure a positive definite covariance matrix. Lastly, we present a novel decomposition of the Generalized Gauss-Newton approximation. Empirically, we show that our Laplacian Metric Learner (LAM) yields well-calibrated uncertainties, reliably detects out-of-distribution examples, and has state-of-the-art predictive performance.

Laplace approximationmetric learninguncertainty quantificationweight posteriorbayesian

BibTeX

@inproceedings{
warburg2023bayesian,
title={Bayesian Metric Learning for Uncertainty Quantification in Image Retrieval},
author={Frederik Rahb{\ae}k Warburg and Marco Miani and Silas Brack and S{\o}ren Hauberg},
booktitle={Thirty-seventh Conference on Neural Information Processing Systems},
year={2023},
url={https://openreview.net/forum?id=58XMiu8kot}
}