Jury Learning: Integrating Dissenting Voices into Machine Learning Models

要旨

Whose labels should a machine learning (ML) algorithm learn to emulate? For ML tasks ranging from online comment toxicity to misinformation detection to medical diagnosis, different groups in society may have irreconcilable disagreements about ground truth labels. Supervised ML today resolves these label disagreements implicitly using majority vote, which overrides minority groups’ labels. We introduce jury learning, a supervised ML approach that resolves these disagreements explicitly through the metaphor of a jury: defining which people or groups, in what proportion, determine the classifier’s prediction. For example, a jury learning model for online toxicity might centrally feature women and Black jurors, who are commonly targets of online harassment. To enable jury learning, we contribute a deep learning architecture that models every annotator in a dataset, samples from annotators’ models to populate the jury, then runs inference to classify. Our architecture enables juries that dynamically adapt their composition, explore counterfactuals, and visualize dissent. A field evaluation finds that practitioners construct diverse juries that alter 14% of classification outcomes.

受賞
Best Paper
著者
Mitchell L.. Gordon
Stanford University, Stanford, California, United States
Michelle S.. Lam
Stanford University, Stanford, California, United States
Joon Sung Park
Stanford University, Palo Alto, California, United States
Kayur Patel
Apple Inc, Seattle, Washington, United States
Jeff Hancock
Stanford University, Stanford, California, United States
Tatsunori Hashimoto
Stanford University, Stanford, California, United States
Michael S.. Bernstein
Stanford University, Stanford, California, United States
論文URL

https://dl.acm.org/doi/abs/10.1145/3491102.3502004

動画

会議: CHI 2022

The ACM CHI Conference on Human Factors in Computing Systems (https://chi2022.acm.org/)

セッション: Agents in the Loop

292
5 件の発表
2022-05-04 18:00:00
2022-05-04 19:15:00