Timezone: »
Spotlight
Pruning has a disparate impact on model accuracy
Cuong Tran · Ferdinando Fioretto · Jung-Eun Kim · Rakshit Naidu
Network pruning is a widely-used compression technique that is able to significantly scale down overparameterized models with minimal loss of accuracy. This paper shows that pruning may create or exacerbate disparate impacts. The paper sheds light on the factors to cause such disparities, suggesting differences in gradient norms and distance to decision boundary across groups to be responsible for this critical issue. It analyzes these factors in detail, providing both theoretical and empirical support, and proposes a simple, yet effective, solution that mitigates the disparate impacts caused by pruning.
Author Information
Cuong Tran (Syracuse University)
Ferdinando Fioretto (Syracuse University)
Jung-Eun Kim (Computer Science, North Carolina State University)
Rakshit Naidu (School of Computer Science, Carnegie Mellon University)
Related Events (a corresponding poster, oral, or spotlight)
-
2022 Poster: Pruning has a disparate impact on model accuracy »
Thu. Dec 1st 05:00 -- 07:00 PM Room Hall J #933
More from the Same Authors
-
2023 Workshop: Algorithmic Fairness through the Lens of Time »
Awa Dieng · Miriam Rateike · Golnoosh Farnadi · Ferdinando Fioretto · Jessica Schrouff -
2022 : Panel »
Ferdinando Fioretto · Amir-Hossein Karimi · Pratyusha Kalluri · Reza Shokri · Elizabeth Watkins · Su Lin Blodgett -
2022 Workshop: Algorithmic Fairness through the Lens of Causality and Privacy »
Awa Dieng · Miriam Rateike · Golnoosh Farnadi · Ferdinando Fioretto · Matt Kusner · Jessica Schrouff -
2021 Poster: Differentially Private Empirical Risk Minimization under the Fairness Lens »
Cuong Tran · My Dinh · Ferdinando Fioretto -
2021 Poster: Learning Hard Optimization Problems: A Data Generation Perspective »
James Kotary · Ferdinando Fioretto · Pascal Van Hentenryck