Identifying Significant Predictive Bias in Classifiers
arXiv:1611.08292
Abstract
We present a novel subset scan method to detect if a probabilistic binary classifier has statistically significant bias -- over or under predicting the risk -- for some subgroup, and identify the characteristics of this subgroup. This form of model checking and goodness-of-fit test provides a way to interpretably detect the presence of classifier bias or regions of poor classifier fit. This allows consideration of not just subgroups of a priori interest or small dimensions, but the space of all possible subgroups of features. To address the difficulty of considering these exponentially many possible subgroups, we use subset scan and parametric bootstrap-based methods. Extending this method, we can penalize the complexity of the detected subgroup and also identify subgroups with high classification errors. We demonstrate these methods and find interesting results on the COMPAS crime recidivism and credit delinquency data.
Presented as a poster at the 2017 Workshop on Fairness, Accountability, and Transparency in Machine Learning (FAT/ML 2017); earlier version presented at NIPS 2016 Workshop on Interpretable Machine Learning in Complex Systems
Cited by in corpus (12)
- Fairness in Machine Learning: A Survey
- Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness
- Distill-and-Compare: Auditing Black-Box Models Using Transparent Model Distillation
- A survey on datasets for fairness-aware machine learning
- Fairer and more accurate, but for whom?
- To Split or Not to Split: The Impact of Disparate Treatment in Classification
- Maximum Weighted Loss Discrepancy
- Automated Supervised Feature Selection for Differentiated Patterns of Care
- Non-Comparative Fairness for Human-Auditing and Its Relation to Traditional Fairness Notions
- Preservation of Anomalous Subgroups On Machine Learning Transformed Data
- On the Identification of Fair Auditors to Evaluate Recommender Systems based on a Novel Non-Comparative Fairness Notion
- Post-discovery Analysis of Anomalous Subsets