Evaluating the Utility of Conformal Prediction Sets for AI-Advised Image Labeling

要旨

As deep neural networks are more commonly deployed in high-stakes domains, their black-box nature makes uncertainty quantification challenging. We investigate the effects of presenting conformal prediction sets---a distribution-free class of methods for generating prediction sets with specified coverage---to express uncertainty in AI-advised decision-making. Through a large online experiment, we compare the utility of conformal prediction sets to displays of Top-$1$ and Top-$k$ predictions for AI-advised image labeling. In a pre-registered analysis, we find that the utility of prediction sets for accuracy varies with the difficulty of the task: while they result in accuracy on par with or less than Top-$1$ and Top-$k$ displays for easy images, prediction sets excel at assisting humans in labeling out-of-distribution (OOD) images, especially when the set size is small. Our results empirically pinpoint practical challenges of conformal prediction sets and provide implications on how to incorporate them for real-world decision-making.

受賞
Honorable Mention
著者
Dongping Zhang
Northwestern University, Evanston, Illinois, United States
Angelos Chatzimparmpas
Northwestern University, Evanston, Illinois, United States
Negar Kamali
Northwestern University, Evanston, Illinois, United States
Jessica Hullman
Northwestern University, Evanston, Illinois, United States
論文URL

https://doi.org/10.1145/3613904.3642446

動画

会議: CHI 2024

The ACM CHI Conference on Human Factors in Computing Systems (https://chi2024.acm.org/)

セッション: Evaluating AI Technologies A

321
5 件の発表
2024-05-15 01:00:00
2024-05-15 02:20:00