A Meta-heuristic Approach to Estimate and Explain Classifier Uncertainty

TL;DR

This paper introduces a model-agnostic meta-heuristic framework that estimates and explains classifier uncertainty, aiding end-users in understanding when models may be unreliable, especially in sensitive domains like healthcare.

Contribution

It proposes novel class-independent meta-heuristics integrated into a meta-learning framework to quantify instance complexity and uncertainty, improving interpretability and risk estimation.

Findings

Outperformed predicted probabilities in identifying risky instances

Provided a model-agnostic approach for uncertainty estimation

Enhanced interpretability of classifier decisions

Abstract

Trust is a crucial factor affecting the adoption of machine learning (ML) models. Qualitative studies have revealed that end-users, particularly in the medical domain, need models that can express their uncertainty in decision-making allowing users to know when to ignore the model's recommendations. However, existing approaches for quantifying decision-making uncertainty are not model-agnostic, or they rely on complex statistical derivations that are not easily understood by laypersons or end-users, making them less useful for explaining the model's decision-making process. This work proposes a set of class-independent meta-heuristics that can characterize the complexity of an instance in terms of factors are mutually relevant to both human and ML decision-making. The measures are integrated into a meta-learning framework that estimates the risk of misclassification. The proposed framework outperformed predicted probabilities in identifying instances at risk of being misclassified. The proposed measures and framework hold promise for improving model development for more complex instances, as well as providing a new means of model abstention and explanation.