# Risk-based Calibration for Generative Classifiers

arXiv: https://arxiv.org/abs/2409.03542 · HTML: https://mlsanity.com/paper/2409.03542/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 31 references that are in the map.
- First posted on arXiv 5 Sep 2024.

## Builds on (most cited first, 7 of 7)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms](https://mlsanity.com/paper/1708.07747/) (2017)
- [Generative Models for Effective ML on Private, Decentralized Datasets](https://mlsanity.com/paper/1911.06679/) (2019)
- [Revisiting Discriminative vs. Generative Classifiers: Theory and Implications](https://mlsanity.com/paper/2302.02334/) (2023)
- [Classification of datasets with imputed missing values: does imputation quality matter?](https://mlsanity.com/paper/2206.08478/) (2022)
- [Probabilistic Imputation for Time-series Classification with Missing Data](https://mlsanity.com/paper/2308.06738/) (2023)
- [Learning Vine Copula Models For Synthetic Data Generation](https://mlsanity.com/paper/1812.01226/) (2018)

## Built on by (most cited first, 2 of 2)

- [Decentralized Federated Learning of Probabilistic Generative Classifiers](https://mlsanity.com/paper/2507.17285/) (2025)
- [Hierarchical Empirical-Bayes Naive Bayes: Minimax Smoothing and Calibration with AODE Extension](https://mlsanity.com/paper/2608.11162/) (2026)

## Abstract (arXiv)

Generative classifiers are constructed on the basis of a joint probability distribution and are typically learned using closed-form procedures that rely on data statistics and maximize scores related to data fitting. However, these scores are not directly linked to supervised classification metrics such as the error, i.e., the expected 0-1 loss. To address this limitation, we propose a learning procedure called risk-based calibration (RC) that iteratively refines the generative classifier by adjusting its joint probability distribution according to the 0-1 loss in training samples. This is achieved by reinforcing data statistics associated with the true classes while weakening those of incorrect classes. As a result, the classifier progressively assigns higher probability to the correct labels, improving its training error. Results on 20 heterogeneous datasets using both na"ive Bayes and quadratic discriminant analysis show that RC significantly outperforms closed-form learning procedures in terms of both training error and generalization error. In this way, RC bridges the gap between traditional generative approaches and learning procedures guided by performance measures, ensuring a closer alignment with supervised classification objectives.