Individual Confidential Computing of Polynomials over Non-Uniform Information
Saar Tarnopolsky 1 , Zirui (Ken) Deng 2 , Vinayak Ramkumar 3 , Netanel Raviv 2 , and Alejandro Cohen 1 · 2025
Read Individual Confidential Computing of Polynomials over Non-Uniform Information by Saar Tarnopolsky 1 , Zirui (Ken) Deng 2 , Vinayak Ramkumar 3 , Netanel Raviv 2 , and Alejandro Cohen 1 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-15645.txt.
Description
Tarnopolsky and co-authors tackle private computing when user data does not arrive evenly spread. Many secure-computing schemes assume uniform data, yet real data is lumpy, and forcing it uniform is hard. Their scheme lets users compute polynomial functions over such data while keeping each person's input hidden.
The trick is linear hashing, which reshapes uneven data into something close to uniform. On top of that, the scheme guards every subset of users at once, not just the whole group. The authors bound the leakage and show it stays negligible, even against the admin running the service and even if workers collude. The guarantee is information-theoretic: it rests on math, not on limits of the attacker's computers.
Catalog record
| CREATOR | Saar Tarnopolsky 1 , Zirui (Ken) Deng 2 , Vinayak Ramkumar 3 , Netanel Raviv 2 , and Alejandro Cohen 1 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 41 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-15645