Summary of Disparisk: Auditing Fairness Through Usable Information, by Jonathan Vasquez et al.
DispaRisk: Auditing Fairness Through Usable Information
by Jonathan Vasquez, Carlotta Domeniconi, Huzefa Rangwala
First submitted to arxiv on: 20 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 A novel framework called DispaRisk is introduced to proactively assess the potential risks of disparities in datasets during the initial stages of a machine learning (ML) pipeline. Recent advancements in usable information theory are leveraged to identify datasets with a high risk of discrimination, detect model families prone to biases within an ML pipeline, and enhance the explainability of these bias risks. The framework’s effectiveness is evaluated by benchmarking it against commonly used datasets in fairness research, demonstrating its capabilities to mitigate societal biases present in datasets. This work contributes to the development of fairer ML systems by providing a robust tool for early bias detection and mitigation. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Machine learning algorithms are used everywhere and have both good and bad effects on society. Sometimes, these algorithms make things worse for certain groups of people, like minorities. To fix this problem, we need to identify biases in the data before we use it to train a model. A new way to do this is called DispaRisk. It uses special math ideas to look at datasets and see if they might be unfair. We tested DispaRisk with some popular datasets and found that it can spot when a dataset has a high chance of being biased or unfair. This helps us make better models that treat everyone fairly. |
Keywords
» Artificial intelligence » Machine learning
