7 papers
Certification of Machine Learning Models via Directional Sharpness
Gefei Tan, Adria Gascon, Sarah Meiklejohn +1
In machine learning, model certification has been identified as an important method for gaining assurance about a model's trustworthiness and quality. A model's quality is largely…
Hardening Confidential Federated Compute against Side-channel Attacks
James Bell-Clark, Albert Cheu, Adria Gascon +1
In this work, we identify a set of side-channels in our Confidential Federated Compute platform that a hypothetical insider could exploit to circumvent differential privacy (DP) gu…
Privacy Reasoning in Ambiguous Contexts
Ren Yi, Octavian Suciu, Adria Gascon +3
We study the ability of language models to reason about appropriate information disclosure - a central aspect of the evolving field of agentic privacy. Whereas previous works have…
SNPeek: Side-Channel Analysis for Privacy Applications on Confidential VMs
Ruiyi Zhang, Albert Cheu, Adria Gascon +4
Confidential virtual machines (CVMs) based on trusted execution environments (TEEs) enable new privacy-preserving solutions. Yet, they leave side-channel leakage outside their thre…
Confidential Federated Computations
Hubert Eichner, Daniel Ramage, Kallista Bonawitz +11
Federated Learning and Analytics (FLA) have seen widespread adoption by technology platforms for processing sensitive on-device data. However, basic FLA systems have privacy limita…
Secure Stateful Aggregation: A Practical Protocol with Applications in Differentially-Private Federated Learning
Marshall Ball, James Bell-Clark, Adria Gascon +3
Recent advances in differentially private federated learning (DPFL) algorithms have found that using correlated noise across the rounds of federated learning (DP-FTRL) yields prova…