paper-with-me

홈 › Papers

Debiasing classifiers: is reality at variance with expectation?

2020-11-04 · Ashrya Agrawal, Florian Pfisterer, Bernd Bischl, Francois Buet-Golfouse, Srijan Sood, Jiahao Chen, Sameena Shah, Sebastian Vollmer

We present an empirical study of debiasing methods for classifiers, showing that debiasers often fail in practice to generalize out-of-sample, and can in fact make fairness worse rather than better. A rigorous evaluation of the debiasing treatment effect requires extensive cross-validation beyond what is usually done. We demonstrate that this phenomenon can be explained as a consequence of bias-variance trade-off, with an increase in variance necessitated by imposing a fairness constraint. Follow-up experiments validate the theoretical prediction that the estimation variance depends strongly on the base rates of the protected class. Considering fairness--performance trade-offs justifies the counterintuitive notion that partial debiasing can actually yield better results in practice on out-of-sample data.

📄 PDF Abstract BibTeX arXiv:2011.02407

Code (0)

등록된 구현이 없습니다.

Tasks

Fairness

Similar Papers 제목 키워드 기반

Ensembling over Classifiers: a Bias-Variance Perspective

2022-06-21 · Neha Gupta, Jamie Smith, Ben Adlam, Zelda Mariet

Ensembles are a straightforward, remarkably effective method for improving the accuracy,calibration, and robustness of models on classification tasks; yet, the reasons that underlie their success remain an active area of…

Adaptive Invariant Extended Kalman Filter with Noise Covariance Tuning for Attitude Estimation

2024-10-02 · Yash Pandey, Rahul Bhattacharyya, Yatindra Nath Singh

Attitude estimation is crucial in aerospace engineering, robotics, and virtual reality applications, but faces difficulties due to nonlinear system dynamics and sensor limitations. This paper addresses the challenge of a…

Unlabeled Debiasing in Downstream Tasks via Class-wise Low Variance Regularization

2024-09-29 · Shahed Masoudian, Markus Frohmann, Navid Rekabsaz, Markus Schedl

Language models frequently inherit societal biases from their training data. Numerous techniques have been proposed to mitigate these biases during both the pre-training and fine-tuning stages. However, fine-tuning a pre…

AttributeLanguage ModelingLanguage Modelling

Leveraging Nested MLMC for Sequential Neural Posterior Estimation with Intractable Likelihoods

2024-01-30 · Xiliang Yang, Yifei Xiong, Zhijian He

Sequential neural posterior estimation (SNPE) techniques have been recently proposed for dealing with simulation-based models with intractable likelihoods. They are devoted to learning the posterior from adaptively propo…

Debiasing Vision-Language Models via Biased Prompts

2023-01-31 · Ching-Yao Chuang, Varun Jampani, Yuanzhen Li, Antonio Torralba 외

Machine learning models have been shown to inherit biases from their training datasets. This can be particularly problematic for vision-language foundation models trained on uncurated datasets scraped from the internet. …