Inspired by cold boot attacks, Heninger and Shacham (Crypto 2009) initiated the study of the problem of how to recover an RSA private key from a noisy version of that key. They gave an algorithm for the case where some bits of the private key are known with certainty. Their ideas were extended by Henecka, May and Meurer (Crypto 2010) to produce an algorithm that works when all the key bits are subject to error. In this paper, we bring a coding-theoretic viewpoint to bear on the problem of noisy RSA key recovery. This viewpoint allows us to cast the previous work as part of a more general framework. In turn, this enables us to explain why the previous algorithms do not solve the motivating cold boot problem, and to design a new algorithm that does (and more). In addition, we are able to use concepts and tools from coding theory-channel capacity, list decoding algorithms, and random coding techniques-to derive bounds on the performance of the previous and our new algorithm.

No abstract

Federated Learning enables a population of clients, working with a trusted server, to collaboratively learn a shared machine learning model while keeping each client's data within its own local systems. This reduces the risk of exposing sensitive data, but it is still possible to reverse engineer information about a client's private data set from communicated model parameters. Most federated learning systems therefore use differential privacy to introduce noise to the parameters. This adds uncertainty to any attempt to reveal private client data, but also reduces the accuracy of the shared model, limiting the useful scale of privacy-preserving noise. A system can further reduce the coordinating server's ability to recover private client information, without additional accuracy loss, by also including secure multiparty computation. An approach combining both techniques is especially relevant to financial firms as it allows new possibilities for collaborative learning without exposing sensitive client data. This could produce more accurate models for important tasks like optimal trade execution, credit origination, or fraud detection. The key contributions of this paper are: We present a privacy-preserving federated learning protocol to a non-specialist audience, demonstrate it using logistic regression on a real-world credit card fraud data set, and evaluate it using an open-source simulation platform which we have adapted for the development of federated learning systems.

No abstract