Muffin: A Framework Toward Multi-Dimension AI Fairness by Uniting Off-the-Shelf Models
It addresses fairness issues in critical applications like healthcare and autonomous driving, focusing on multiple attributes rather than single ones, though it is incremental as it builds on existing models.
The paper tackles the problem of multi-dimensional fairness in AI models, where optimizing fairness on one attribute can degrade others, and proposes the Muffin framework to unite off-the-shelf models for simultaneous improvements, achieving fairness gains of 26.32% and 20.37% on two attributes with a 5.58% accuracy gain.
Model fairness (a.k.a., bias) has become one of the most critical problems in a wide range of AI applications. An unfair model in autonomous driving may cause a traffic accident if corner cases (e.g., extreme weather) cannot be fairly regarded; or it will incur healthcare disparities if the AI model misdiagnoses a certain group of people (e.g., brown and black skin). In recent years, there have been emerging research works on addressing unfairness, and they mainly focus on a single unfair attribute, like skin tone; however, real-world data commonly have multiple attributes, among which unfairness can exist in more than one attribute, called 'multi-dimensional fairness'. In this paper, we first reveal a strong correlation between the different unfair attributes, i.e., optimizing fairness on one attribute will lead to the collapse of others. Then, we propose a novel Multi-Dimension Fairness framework, namely Muffin, which includes an automatic tool to unite off-the-shelf models to improve the fairness on multiple attributes simultaneously. Case studies on dermatology datasets with two unfair attributes show that the existing approach can achieve 21.05% fairness improvement on the first attribute while it makes the second attribute unfair by 1.85%. On the other hand, the proposed Muffin can unite multiple models to achieve simultaneously 26.32% and 20.37% fairness improvement on both attributes; meanwhile, it obtains 5.58% accuracy gain.