Revisiting Precision and Recall Definition for Generative Model Evaluation

By Loïc Simon, Ryan Webster, Julien Rabin. 2019

20 citations recorded by the source corpus.

In this article we revisit the definition of Precision-Recall (PR) curves for generative models proposed by Sajjadi et al. (arXiv:1806.00035). Rather than providing a scalar for generative quality, PR curves distinguish mode-collapse (poor recall) and bad quality (poor precision). We first generalize their formulation to arbitrary measures, hence removing any restriction to finite support. We also expose a bridge between PR curves and type I and type II error rates of likelihood ratio classifiers on the task of discriminating between samples of the two distributions. Building upon this new perspective, we propose a novel algorithm to approximate precision-recall curves, that shares some interesting methodological properties with the hypothesis testing technique from Lopez-Paz et al (arXiv:1610.06545). We demonstrate the interest of the proposed formulation over the original approach on controlled multi-modal datasets.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Generative Models, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Samuli Laine
2. Timo Aila
3. Dan Friedman
4. Tuomas Kynkaanniemi
5. Tero Karras
6. Adji Bousso Dieng
7. Jaakko Lehtinen
8. Guo-Jun Qi

In Computer Vision and Pattern Recognition

1. Luc Van Gool
2. Ming-Hsuan Yang
3. Dacheng Tao
4. Antonio Torralba
5. Chen Change Loy
6. Trevor Darrell
7. Xiaogang Wang
8. Andrea Vedaldi

In Generative Models

1. Tero Karras
2. Eli Shechtman
3. Richard Zhang
4. Timo Aila
5. Jun-Yan Zhu
6. Daniel Cohen-Or
7. Jaakko Lehtinen
8. Samuli Laine

In Evaluation & Benchmarks

1. Tero Karras
2. Timo Aila
3. Adji Bousso Dieng
4. Samuli Laine
5. Dan Friedman
6. Jaakko Lehtinen
7. Richard Zhang
8. Prafulla Dhariwal

Citations in the corpus

• Improved Precision and Recall Metric for Assessing Generative Models (2019, background): “[26] extend Sajjadi’s formulation to arbitrary probability distributions and provide a practical algorithm that estimates precision and recall by training a post hoc classifier.”
• The Vendi Score: A Diversity Evaluation Metric for Machine Learning (2022, background): “Several other variations of precision and recall have been proposed (Kynkäänniemi et al., 2019; Simon et al., 2019; Naeem et al., 2020).”