6 papers · 1 filter
Certified Robustness to Data Poisoning in Gradient-Based Training
Philip Sosnin, Mark N. Müller, Maximilian Baader +2
Modern machine learning pipelines leverage large amounts of public data, making it infeasible to guarantee data quality and leaving models open to poisoning and backdoor attacks. P…
Variational Routing: A Scalable Bayesian Framework for Calibrated Mixture-of-Experts Transformers
Albus Yizhuo Li, Matthew Wicker
Foundation models are increasingly being deployed in contexts where understanding the uncertainty of their outputs is critical to ensuring responsible deployment. While Bayesian me…
SafeAdapt: Provably Safe Policy Updates in Deep Reinforcement Learning
Maksim Anisimov, Francesco Belardinelli, Matthew Wicker
Safety guarantees are a prerequisite to the deployment of reinforcement learning (RL) agents in safety-critical tasks. Often, deployment environments exhibit non-stationary dynamic…
Provably Safe Model Updates
Leo Elmecker-Plakolm, Pierre Fasterling, Philip Sosnin +2
Safety-critical environments are inherently dynamic. Distribution shifts, emerging vulnerabilities, and evolving requirements demand continuous updates to machine learning models.…
Abstract Gradient Training: A Unified Certification Framework for Data Poisoning, Unlearning, and Differential Privacy
Philip Sosnin, Matthew Wicker, Josh Collyer +1
The impact of inference-time data perturbation (e.g., adversarial attacks) has been extensively studied in machine learning, leading to well-established certification techniques fo…
Certification for Differentially Private Prediction in Gradient-Based Training
Matthew Wicker, Philip Sosnin, Igor Shilov +5
We study private prediction where differential privacy is achieved by adding noise to the outputs of a non-private model. Existing methods rely on noise proportional to the global…