Silva: Interactively Assessing Machine Learning Fairness Using Causality
Authors
Machine learning models risk encoding unfairness on the part of their developers or data sources. However, assessing fairness is challenging as analysts might misidentify sources of bias, fail to notice them, or misapply metrics. In this paper we introduce Silva, a system for exploring potential sources of unfairness in datasets or machine learning models interactively. Silva directs user attention to relationships between attributes through a global causal view, provides interactive recommendations, presents intermediate results, and visualizes metrics. We describe the implementation of Silva, identify salient design and technical challenges, and provide an evaluation of the tool in comparison to an existing fairness optimization tool.
Research Questions / Practical Problems
Question signals indexed for this paper.
- 67%
Towards Fairness in Practice: A Practitioner-Oriented Rubric for Evaluating Fair ML Toolkits
CHI '21· AI Ethics, Fairness & Accountability +1
- 67%
Jury Learning: Integrating Dissenting Voices into Machine Learning Models
CHI '22· AI Ethics, Fairness & Accountability +1
- 67%
Capable but Amoral? Comparing AI and Human Expert Collaboration in Ethical Decision Making
CHI '22· AI Ethics, Fairness & Accountability +1
- 67%
Out of Context: Investigating the Bias and Fairness Concerns of "Artificial Intelligence as a Service"
CHI '23· AI Ethics, Fairness & Accountability +1
- 67%
“It is currently hodgepodge”: Examining AI/ML Practitioners’ Challenges during Co-production of Responsible AI Values
CHI '23· AI Ethics, Fairness & Accountability +1
- 67%
STILE: Exploring and Debugging Social Biases in Pre-trained Text Representations
CHI '24· AI Ethics, Fairness & Accountability +1
- 63%
Exploring the Association between Moral Foundations and Judgements of AI Behaviour
CHI '24· AI Ethics, Fairness & Accountability +2
Based on Jaccard similarity of research subtopics & professions (≥60%)