Summary of Privacy-preserving Orthogonal Aggregation For Guaranteeing Gender Fairness in Federated Recommendation, by Siqing Zhang et al.
Privacy-Preserving Orthogonal Aggregation for Guaranteeing Gender Fairness in Federated Recommendation
by Siqing Zhang, Yuchen Ding, Wei Tang, Wei Sun, Yong Liao, Peng Yuan Zhou
First submitted to arxiv on: 29 Nov 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 The proposed Privacy-Preserving Orthogonal Aggregation (PPOA) method tackles the issue of group fairness in federated recommendation systems, particularly with regards to gender fairness. Current state-of-the-art methods only focus on one aspect of the problem, neglecting others that can lead to detrimental effects on model training. PPOA employs a secure aggregation scheme and quantization technique to prevent suppression of minority groups by the majority and preserve distinct preferences for better group fairness. This approach achieves optimal fairness in most cases, leading to up to 8.25% and 6.36% improvement in recommendation effectiveness for females and males, respectively. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Federated recommendation systems need to be fair and private! Currently, they can’t ensure that different groups get the right recommendations. The problem is even harder when we’re dealing with sensitive information like gender. This paper proposes a new way to make these systems work better by preserving privacy and fairness. They test their approach on real-world datasets and show that it improves recommendation accuracy for both women and men. |
Keywords
» Artificial intelligence » Quantization
