# Improving CNN classifiers by estimating test-time priors

arXiv: https://arxiv.org/abs/1805.08235 · HTML: https://mlsanity.com/paper/1805.08235/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 May 2018.

## Built on by (most cited first, 7 of 7)

- [A Comprehensive Survey on Test-Time Adaptation under Distribution Shifts](https://mlsanity.com/paper/2303.15361/) (2023)
- [Danish Fungi 2020 -- Not Just Another Image Recognition Dataset](https://mlsanity.com/paper/2103.10107/) (2021)
- [Improving Zero-Shot Models with Label Distribution Priors](https://mlsanity.com/paper/2212.00784/) (2022)
- [The Hitchhiker's Guide to Prior-Shift Adaptation](https://mlsanity.com/paper/2106.11695/) (2021)
- [Contrastive Classification and Representation Learning with Probabilistic Interpretation](https://mlsanity.com/paper/2211.03646/) (2022)
- [GLAMI-1M: A Multilingual Image-Text Fashion Dataset](https://mlsanity.com/paper/2211.14451/) (2022)
- [Adapting Classifiers To Changing Class Priors During Deployment](https://mlsanity.com/paper/2309.01357/) (2023)

## Abstract (arXiv)

The problem of different training and test set class priors is addressed in the context of CNN classifiers. We compare two different approaches to estimating the new priors: an existing Maximum Likelihood Estimation approach (optimized by an EM algorithm or by projected gradient descend) and a proposed Maximum a Posteriori approach, which increases the stability of the estimate by introducing a Dirichlet hyper-prior on the class prior probabilities. Experimental results show a significant improvement on the fine-grained classification tasks using known evaluation-time priors, increasing the top-1 accuracy by 4.0% on the FGVC iNaturalist 2018 validation set and by 3.9% on the FGVCx Fungi 2018 validation set. Estimation of the unknown test set priors noticeably increases the accuracy on the PlantCLEF dataset, allowing a single CNN model to achieve state-of-the-art results and outperform the competition-winning ensemble of 12 CNNs. The proposed Maximum a Posteriori estimation increases the prediction accuracy by 2.8% on PlantCLEF 2017 and by 1.8% on FGVCx Fungi, where the existing MLE method would lead to a decrease accuracy.