Scholar
Mathias Lécuyer
Google Scholar ID: WeIvMTUAAAAJ
University of British Columbia
Citations & Impact
Citations
1,708
H-index
13
i10-index
14
Publications
20
Co-authors
55
list available
Contact
Publications
Resume (English only)
Academic Achievements
- Published multiple papers, including but not limited to:
- - Adaptive Diffusion Denoised Smoothing: Certified Robustness via Randomized Smoothing with Differentially Private Guided Denoising Diffusion
- - On the Performance of Differentially Private Optimization with Heavy-Tail Class Imbalance
- - Connecting Thompson Sampling and UCB: Towards More Efficient Trade-offs Between Privacy and Regret
- - FedFetch: Faster Federated Learning with Adaptive Downstream Prefetching
- - DPack: Efficiency-Oriented Privacy Budget Scheduling
- - Training and Evaluating Causal Forecasting Models for Time-Series
- - Adaptive Randomized Smoothing: Certified Adversarial Robustness for Multi-Step Defences
- - PANORAMIA: Privacy Auditing of Machine Learning Models without Retraining
- - Cookie Monster: Efficient On-Device Budgeting for Differentially-Private Ad-Measurement Systems
Research Experience
- Assistant Professor at the University of British Columbia, involved in research with Systopia, UBC S&P, TrustML, and CAIDA. Former postdoctoral researcher at Microsoft Research (NY).
Education
- PhD from Columbia University; Postdoctoral researcher at Microsoft Research (NY).
Background
- Currently an assistant professor at the University of British Columbia, focusing on trustworthy Artificial Intelligence (AI) systems, particularly on auditing AI models and developing techniques to enforce provable guarantees in models and their data ecosystems.
Miscellany
- Teaches courses such as Causal Machine Learning, Machine Learning and Data Mining, Applied Machine Learning, and Differential Privacy - Theory and Practice.
