Enhancing Monte Carlo Dropout Performance for Uncertainty Quantification

TL;DR

This paper improves Monte Carlo Dropout for uncertainty quantification in deep neural networks by integrating optimization techniques and an uncertainty-aware loss, leading to better calibration and reliability in high-stakes applications.

Contribution

The paper introduces novel frameworks combining GWO, BO, PSO, and an uncertainty-aware loss to enhance MCD's uncertainty estimates in deep learning models.

Findings

Outperforms baseline MCD by 2-3% in accuracy and uncertainty calibration

Achieves significantly better uncertainty calibration across multiple datasets

Demonstrates improved trustworthiness of deep models in critical applications

Abstract

Knowing the uncertainty associated with the output of a deep neural network is of paramount importance in making trustworthy decisions, particularly in high-stakes fields like medical diagnosis and autonomous systems. Monte Carlo Dropout (MCD) is a widely used method for uncertainty quantification, as it can be easily integrated into various deep architectures. However, conventional MCD often struggles with providing well-calibrated uncertainty estimates. To address this, we introduce innovative frameworks that enhances MCD by integrating different search solutions namely Grey Wolf Optimizer (GWO), Bayesian Optimization (BO), and Particle Swarm Optimization (PSO) as well as an uncertainty-aware loss function, thereby improving the reliability of uncertainty quantification. We conduct comprehensive experiments using different backbones, namely DenseNet121, ResNet50, and VGG16, on various datasets, including Cats vs. Dogs, Myocarditis, Wisconsin, and a synthetic dataset (Circles). Our proposed algorithm outperforms the MCD baseline by 2-3% on average in terms of both conventional accuracy and uncertainty accuracy while achieving significantly better calibration. These results highlight the potential of our approach to enhance the trustworthiness of deep learning models in safety-critical applications.