Reliable uncertainty estimate for antibiotic resistance classification with Stochastic Gradient Langevin Dynamics

TL;DR

This paper introduces a deep learning model trained with Stochastic Gradient Langevin Dynamics to improve uncertainty estimation in antibiotic resistance classification, especially for out-of-distribution data, addressing limitations of traditional training methods.

Contribution

It presents a novel application of SGLD for antibiotic resistance classification, enhancing uncertainty estimation over standard optimization algorithms.

Findings

SGLD-trained model provides better OoD uncertainty estimates.

Improved robustness in antibiotic resistance classification.

Demonstrates advantages over Adam in uncertainty quantification.

Abstract

Antibiotic resistance monitoring is of paramount importance in the face of this on-going global epidemic. Deep learning models trained with traditional optimization algorithms (e.g. Adam, SGD) provide poor posterior estimates when tested against out-of-distribution (OoD) antibiotic resistant/non-resistant genes. In this paper, we introduce a deep learning model trained with Stochastic Gradient Langevin Dynamics (SGLD) to classify antibiotic resistant genes. The model provides better uncertainty estimates when tested against OoD data compared to traditional optimization methods such as Adam.