Explainability through uncertainty: Trustworthy decision-making with neural networks

Abstract

Uncertainty is a key feature of any machine learning model and is particularly important in neural networks, which tend to be overconfident. This overconfidence is worrying under distribution shifts, where the model performance silently degrades as the data distribution diverges from the training data distribution. Uncertainty estimation offers a solution to overconfident models, communicating when the output should (not) be trusted. Although methods for uncertainty estimation have been developed, they have not been explicitly linked to the field of explainable artificial intelligence (XAI). Furthermore, literature in operations research ignores the actionability component of uncertainty estimation and does not consider distribution shifts. This work proposes a general uncertainty framework, with contributions being threefold: (i) uncertainty estimation in ML models is positioned as an XAI technique, giving local and model-specific explanations; (ii) classification with rejection is used to reduce misclassifications by bringing a human expert in the loop for uncertain observations; (iii) the framework is applied to a case study on neural networks in educational data mining subject to distribution shifts. Uncertainty as XAI improves the model's trustworthiness in downstream decision-making tasks, giving rise to more actionable and robust machine learning systems in operations research.

arXiv commentAccepted Manuscript version of an article published in the European Journal of Operational Research

References in corpus (7)

• Towards A Rigorous Science of Interpretable Machine Learning
• Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methods
• Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift
• Student Success Prediction in MOOCs
• Forecasting remaining useful life: Interpretable deep learning approach via variational Bayesian inferences
• Learning Uncertainty with Artificial Neural Networks for Improved Predictive Process Monitoring
• Predicting student performance using sequence classification with time-based windows

Cited by in corpus (1)

• Fast and reliable uncertainty quantification with neural network ensembles for industrial image classification