Summary of Improving Fairness in Credit Lending Models Using Subgroup Threshold Optimization, by Cecilia Ying et al.
Improving Fairness in Credit Lending Models using Subgroup Threshold Optimization
by Cecilia Ying, Stephen Thomas
First submitted to arxiv on: 15 Mar 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Risk Management (q-fin.RM)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes a new fairness technique, Subgroup Threshold Optimizer (STO), to improve the accuracy of credit lending decisions while reducing bias and unfairness towards certain subgroups. By optimizing classification thresholds for individual subgroups, STO minimizes the overall discrimination score between them. The authors demonstrate the effectiveness of STO on a real-world credit lending dataset, achieving over 90% reduction in gender discrimination. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps make sure that credit decisions are fair and don’t favor certain groups. Right now, many financial companies use predictions from machine learning models to decide who gets loans. But these predictions can be biased, which means they might not be equal for everyone. To fix this problem, some techniques have been developed to remove the bias. The new technique called STO is special because it works without changing the data or the algorithm used. It just adjusts the way predictions are made so that everyone has a fair chance. |
Keywords
* Artificial intelligence * Classification * Machine learning
