An In-Depth Examination of Risk Assessment in Multi-Class Classification Algorithms

TL;DR

This paper evaluates different risk assessment methods in multi-class classification, focusing on calibration and conformal prediction techniques, to improve error probability estimates in safety-critical applications.

Contribution

It introduces a novel conformal prediction-based risk assessment approach that is model-agnostic, easy to implement, and effective across various datasets and models.

Findings

Conformal prediction provides reliable risk estimates across models.

Calibration techniques improve probability accuracy but vary by dataset.

The conformal approach is simple and broadly applicable.

Abstract

Advanced classification algorithms are being increasingly used in safety-critical applications like health-care, engineering, etc. In such applications, miss-classifications made by ML algorithms can result in substantial financial or health-related losses. To better anticipate and prepare for such losses, the algorithm user seeks an estimate for the probability that the algorithm miss-classifies a sample. We refer to this task as the risk-assessment. For a variety of models and datasets, we numerically analyze the performance of different methods in solving the risk-assessment problem. We consider two solution strategies: a) calibration techniques that calibrate the output probabilities of classification models to provide accurate probability outputs; and b) a novel approach based upon the prediction interval generation technique of conformal prediction. Our conformal prediction based approach is model and data-distribution agnostic, simple to implement, and provides reasonable results for a variety of use-cases. We compare the different methods on a broad variety of models and datasets.