一句话：提出 FEDORA，用受控隐私的 ORAM 在联邦学习中高效训练推荐模型。
训练高质量推荐模型需收集敏感用户数据，但联邦学习因推荐模型嵌入表巨大而难以实用。本文提出 FEDORA：每个用户仅基于私有数据下载、训练并上传大表的一小部分，并用 ORAM 隐藏访问模式。为降低 ORAM 的延迟与内存开销，FEDORA 引入 ε-FDP 来形式化权衡隐私与性能，将大型 ORAM 部署在低成本 SSD 上并做 SSD 友好优化，同时支持现代联邦学习运行模式。评估显示，FEDORA 利用私有特征训练仍保持高准确率，相比基线最高降低 24× 延迟并提升超 1000× SSD 寿命，平均降低 5× 延迟、提升 158× SSD 寿命。
Training high-quality recommendation models requires collecting sensitive user data. The popular privacy-enhancing training method, federated learning (FL), cannot be used practically due to these models' large embedding tables. This paper introduces FEDORA, a system for training recommendation models with FL. FEDORA allows each user to only download, train, and upload a small subset of the large tables based on their private data, while hiding the access pattern using oblivious memory (ORAM). FEDORA reduces the ORAM's prohibitive latency and memory overheads by (1) introducing ε-FDP, a formal way to balance the ORAM's privacy with performance, and (2) placing the large ORAM in a power- and cost-efficient SSD with SSD-friendly optimizations. Additionally, FEDORA is carefully designed to support (3) modern operation modes of FL. FEDORA achieves high model accuracy by using private features during training while achieving up to 24× latency and over 1000× SSD lifetime improvement over the baseline. FEDORA achieves high model accuracy by using private features during training while achieving, on average, 5× latency and 158× SSD lifetime improvement over the baseline.
@inproceedings{Liu0SM25,
title = {Practical Federated Recommendation Model Learning Using ORAM with Controlled Privacy},
author = {Jinyu Liu and Wenjie Xiong and G. Edward Suh and Kiwan Maeng},
booktitle = {ASPLOS},
year = {2025},
doi = {10.1145/3676641.3716014}
}