Recycling Privileged Learning and Distribution Matching for Fairness

By Novi Quadrianto, Viktoriia Sharmanska. 2017

87 citations recorded by the source corpus.

Subfields: Safety, Robustness & Interpretability

How researchers see this paper

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

In Safety, Robustness & Interpretability

Citations in the corpus

Exposing and Correcting the Gender Bias in Image Captioning Datasets and Models (2019, methodology): “With the rise of awareness about dataset bias, several frameworks for fairness have been introduced, including [Hardt et al., 2016], [Dwork et al., 2012], [Quadrianto and Sharmanska, 2017], [Goyal et al., 2017].”
Women also Snowboard: Overcoming Bias in Captioning Models (2018, background): “, age, gender, sexual orientation) in a fair manner is increasingly important as AI continues to become more ubiquitous [14,15,16,17].”
Empirical Risk Minimization under Fairness Constraints (2018, background): “Other works tries to match the second order moment [20] or potentially infinitely many moments [37] but these approaches result in non-convex approaches.”
Algorithmic Fairness (2020, background): “[113] suggest using the notion of privileged learning 1 for improving performance in cases where the sensitive information is available at training time but not at testing time.”
Classification with Fairness Constraints: A Meta-Algorithm with Provable Guarantees (2018, background): “[3, 56] also provide a general framework to handle multiple fairness constraints.”
A Review of Domain Adaptation without Target Labels (2019, background): “To tackle this kind of unwanted behavior in algorithms, fairness has to be built into the learning process [68].”
A Review on Fairness in Machine Learning (2022, background): “Quadrianto and Sharmanska [169] suggest using the notion of privileged learning 2 for improving performance in cases where the sensitive information is available at training time but not at testing time.”
Fairness Constraints: A Flexible Approach for Fair Classification (2019, methodology): “However, their works are only limited to a single specific loss function (Goh et al., 2016; Quadrianto and Sharmanska, 2017) or to a single notion of unfairness (Woodworth et al., 2017).”
Fairness in Machine Learning (2020, methodology): “…to the objective [5, 14, 22, 44, 46, 56, 58, 62, 73, 74, 75, 79, 82, 87, 88, 89, 93, 94, 95, 98, 104, 115, 119, 123, 133, 134, 135, 138, 139, 154, 165, 176, 179, 180, 182], others use adversary techniques to maximize the system ability to predict the target while minimizing the ability to…”
Trustworthy Artificial Intelligence: A Review (2022, background): “Decrease F [10, 27, 33, 38, 97, 128, 172] F, P [63, 91, 169, 189] F, A [156]”