Tiziano Santoro
Authored Publications
Sort By
Confidential Federated Computations
Hubert Eichner
Dzmitry Huba
Brett McLarnon
Timon Van Overveldt
Nova Fallen
Albert Cheu
Katharine Daly
Adria Gascon
Marco Gruteser
ArXiv (2024)
Preview abstract
Federated Learning and Analytics (FLA) have seen widespread adoption by technology platforms for processing sensitive on-device data. However, basic FLA systems have privacy limitations: they do not necessarily require anonymization mechanisms like differential privacy (DP), and provide limited protections against a potentially malicious service provider. Adding DP to a basic FLA system currently requires either adding excessive noise to each device's updates, or assuming an honest service provider that correctly implements the mechanism and only uses the privatized outputs. Secure multiparty computation (SMPC) -based oblivious aggregations can limit the service provider's access to individual user updates and improve DP tradeoffs, but the tradeoffs are still suboptimal, and they suffer from scalability challenges and susceptibility to Sybil attacks. This paper introduces a novel system architecture that leverages trusted execution environments (TEEs) and open-sourcing to both ensure confidentiality of server-side computations and provide externally verifiable privacy properties, bolstering the robustness and trustworthiness of private federated computations.
View details
Policy Transparency: Authorization Logic Meets General Transparency to Prove Software Supply Chain Integrity
Andrew Ferraiuolo
Razieh Behjati
ACM Workshop on Software Supply Chain Offensive Research and Ecosystem Defenses, Association for Computing Machinery (2022)
Preview abstract
Building reliable software is challenging because today’s software supply chains are built
and secured from tools and individuals from a broad range of organizations
with complex trust relationships.
In this setting, tracking the origin of each piece of software and understanding the security
and privacy implications of using it is essential. In this work we aim to secure software
supply chains by using verifiable policies in which the origin of information and the
trust assumptions are first-order concerns and abusive evidence is discoverable.
To do so, we propose Policy Transparency, a new paradigm in which
policies are based on authorization logic and all claims issued in this policy
language are made transparent by inclusion in a transparency log. Achieving this
goal in a real-world setting is non-trivial and to do so we propose a novel
software architecture called PolyLog. We find that this combination
of authorization logic and transparency logs is mutually beneficial --
transparency logs allow authorization logic claims to be widely available aiding
in discovery of abuse, and making claims interpretable
with policies allows misbehavior captured in the transparency logs to be
handled proactively.
View details
