DeepAI AI Chat
Log In Sign Up

Feature-Wise Bias Amplification

by   Klas Leino, et al.

We study the phenomenon of bias amplification in classifiers, wherein a machine learning model learns to predict classes with a greater disparity than the underlying ground truth. We demonstrate that bias amplification can arise via an inductive bias in gradient descent methods that results in the overestimation of the importance of moderately-predictive "weak" features if insufficient training data is available. This overestimation gives rise to feature-wise bias amplification -- a previously unreported form of bias that can be traced back to the features of a trained model. Through analysis and experiments, we show that while some bias cannot be mitigated without sacrificing accuracy, feature-wise bias amplification can be mitigated through targeted feature selection. We present two new feature selection algorithms for mitigating bias amplification in linear models, and show how they can be adapted to convolutional neural networks efficiently. Our experiments on synthetic and real data demonstrate that these algorithms consistently lead to reduced bias without harming accuracy, in some cases eliminating predictive bias altogether while providing modest gains in accuracy.


page 1

page 2

page 3

page 4


DIWIFT: Discovering Instance-wise Influential Features for Tabular Data

Tabular data is one of the most common data storage formats in business ...

Feature Selection Methods for Uplift Modeling

Uplift modeling is a predictive modeling technique that estimates the us...

Feature Selection Using Batch-Wise Attenuation and Feature Mask Normalization

Feature selection is generally used as one of the most important pre-pro...

Detecting Racial Bias in Jury Selection

To support the 2019 U.S. Supreme Court case "Flowers v. Mississippi", AP...

Technical Note: Bias and the Quantification of Stability

Research on bias in machine learning algorithms has generally been conce...

REPAIR: Removing Representation Bias by Dataset Resampling

Modern machine learning datasets can have biases for certain representat...

Autocalibration and Tweedie-dominance for Insurance Pricing with Machine Learning

Boosting techniques and neural networks are particularly effective machi...