Summary of Counterfactual Fairness Through Transforming Data Orthogonal to Bias, by Shuyi Chen et al.
Counterfactual Fairness through Transforming Data Orthogonal to Bias
by Shuyi Chen, Shixiang Zhu
First submitted to arxiv on: 26 Mar 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Machine Learning (stat.ML)
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 The proposed Orthogonal to Bias (OB) algorithm is designed to eliminate the influence of continuous sensitive variables in machine learning applications, promoting counterfactual fairness. This model-agnostic approach is based on the assumption of a jointly normal distribution within a structural causal model and demonstrates that counterfactual fairness can be achieved by ensuring data is orthogonal to observed sensitive variables. The OB algorithm includes a sparse variant for improved numerical stability through regularization. Empirical evaluations on simulated and real-world datasets, including settings with both discrete and continuous sensitive variables, show the methodology effectively promotes fairer outcomes without compromising accuracy. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A machine learning model can sometimes treat different groups unfairly. To fix this, researchers developed an algorithm called Orthogonal to Bias (OB). It helps remove the impact of certain factors that might make decisions biased. The algorithm is based on a specific idea about how data relates to these factors and shows that fairer decisions can be made by making sure the data doesn’t include information about sensitive variables. This approach works with many different types of machine learning models and tasks. It also has a special version that makes it more stable and accurate. |
Keywords
* Artificial intelligence * Machine learning * Regularization
