Qi Lei
Scholar

Qi Lei

Google Scholar ID: kGOgaowAAAAJ
Assistant Professor of Mathematics and Data Science, New York University
Machine LearningOptimizationDeep Learning
Citations & Impact
All-time
Citations
2,812
 
H-index
25
 
i10-index
37
 
Publications
20
 
Co-authors
19
list available
Resume (English only)
Academic Achievements
  • Publications:
  • - Paper accepted at UAI 2025: Beyond Invisibility: Learning Robust Visible Watermarks for Stronger Copyright Protection
  • - Paper accepted at ICML 2025: Discrepancies are Virtue: Weak-to-Strong Generalization through Lens of Intrinsic Dimension
  • - Three papers accepted at AISTATS 2025
  • - Paper accepted at ICLR 2025: Beyond Interpretability: The Gains of Feature Monosemanticity on Model Robustness
  • - Two papers accepted at NeurIPS 2024
  • Invited Talks:
  • - Invited talk at Maryland Numerical Analysis Group, the Inaugural Workshop on Frontiers in Statistical Machine Learning, Inverse Methods for Complex Systems under Uncertainty Workshop
  • - Invited talk at IMS@NUS, ICSDS, Harvard Statistics
  • Organized Events:
  • - Organized the minisymposium “Efficient Computation and Learning with Randomized Sampling and Pruning” at SIAM MDS 2024
Research Experience
  • Serves as an Assistant Professor at the Courant Institute of Mathematical Sciences and Center for Data Science, New York University; involved in multiple research projects including weak-to-strong generalization, data reconstruction attack and defense, data and model pruning, and theoretical foundations of pre-trained models.
Education
  • Information not provided
Background
  • Research Interests: Bridging the theoretical and empirical boundary of modern machine learning algorithms, AI safety, data privacy, distributionally robust algorithms, sample- and parameter-efficient learning. Bio: Assistant Professor of Mathematics and Data Science, and by courtesy, Assistant Professor of Computer Science at Courant Institute of Mathematical Sciences and Center for Data Science, New York University. Member of CILVR lab, Math and Data, and Google DeepMind Faculty.
Miscellany
  • Personal interests and hobbies not mentioned