Saved in:
Bibliographic Details
Main Authors: Zhang, Jiaxin, Das, Kamalika, Kumar, Sricharan
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.12664
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911780173774848
author Zhang, Jiaxin
Das, Kamalika
Kumar, Sricharan
author_facet Zhang, Jiaxin
Das, Kamalika
Kumar, Sricharan
contents Uncertainty estimation is a crucial aspect of deploying dependable deep learning models in safety-critical systems. In this study, we introduce a novel and efficient method for deterministic uncertainty estimation called Discriminant Distance-Awareness Representation (DDAR). Our approach involves constructing a DNN model that incorporates a set of prototypes in its latent representations, enabling us to analyze valuable feature information from the input data. By leveraging a distinction maximization layer over optimal trainable prototypes, DDAR can learn a discriminant distance-awareness representation. We demonstrate that DDAR overcomes feature collapse by relaxing the Lipschitz constraint that hinders the practicality of deterministic uncertainty methods (DUMs) architectures. Our experiments show that DDAR is a flexible and architecture-agnostic method that can be easily integrated as a pluggable layer with distance-sensitive metrics, outperforming state-of-the-art uncertainty estimation methods on multiple benchmark problems.
format Preprint
id arxiv_https___arxiv_org_abs_2402_12664
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Discriminant Distance-Aware Representation on Deterministic Uncertainty Quantification Methods
Zhang, Jiaxin
Das, Kamalika
Kumar, Sricharan
Machine Learning
Uncertainty estimation is a crucial aspect of deploying dependable deep learning models in safety-critical systems. In this study, we introduce a novel and efficient method for deterministic uncertainty estimation called Discriminant Distance-Awareness Representation (DDAR). Our approach involves constructing a DNN model that incorporates a set of prototypes in its latent representations, enabling us to analyze valuable feature information from the input data. By leveraging a distinction maximization layer over optimal trainable prototypes, DDAR can learn a discriminant distance-awareness representation. We demonstrate that DDAR overcomes feature collapse by relaxing the Lipschitz constraint that hinders the practicality of deterministic uncertainty methods (DUMs) architectures. Our experiments show that DDAR is a flexible and architecture-agnostic method that can be easily integrated as a pluggable layer with distance-sensitive metrics, outperforming state-of-the-art uncertainty estimation methods on multiple benchmark problems.
title Discriminant Distance-Aware Representation on Deterministic Uncertainty Quantification Methods
topic Machine Learning
url https://arxiv.org/abs/2402.12664