Philip Sosnin

According to our database, Philip Sosnin authored at least 7 papers between 2024 and 2026.

Collaborative distances:

• Dijkstra number of four.
• Erdős number of four.

2026

Exact Certification of Data-Poisoning Attacks Using Mixed-Integer Programming.

Philip Sosnin, Jodie Knapp, Fraser Kennedy, Josh Collyer, Calvin Tsay

CoRR, February, 2026

2025

Provably Safe Model Updates.

Leo Elmecker-Plakolm, Pierre Fasterling, Philip Sosnin, Calvin Tsay, Matthew Wicker

CoRR, December, 2025

Abstract Gradient Training: A Unified Certification Framework for Data Poisoning, Unlearning, and Differential Privacy.

Philip Sosnin, Matthew Wicker, Josh Collyer, Calvin Tsay

CoRR, November, 2025

Certified Robustness to Data Poisoning in Gradient-Based Training.

Philip Sosnin, Mark Niklas Müller, Maximilian Baader, Calvin Tsay, Matthew Wicker

Trans. Mach. Learn. Res., 2025

Certification for Differentially Private Prediction in Gradient-Based Training.

Matthew Wicker, Philip Sosnin, Igor Shilov, Adrianna Janik, Mark Niklas Müller, Yves-Alexandre de Montjoye, Adrian Weller, Calvin Tsay

Proceedings of the Forty-second International Conference on Machine Learning, 2025

2024

Certificates of Differential Privacy and Unlearning for Gradient-Based Training.

Matthew Wicker, Philip Sosnin, Adrianna Janik, Mark Niklas Müller, Adrian Weller, Calvin Tsay

CoRR, 2024

Scaling Mixed-Integer Programming for Certification of Neural Network Controllers Using Bounds Tightening.

Philip Sosnin, Calvin Tsay

Proceedings of the 63rd IEEE Conference on Decision and Control, 2024