Research interests:
Statistical learning theory
Stochastic optimization
High-dimensional statistics
Research interests:
Statistical learning theory
Stochastic optimization
High-dimensional statistics
Publications:
Parallelized Midpoint Randomization for Langevin Monte Carlo
L. Yu, A.S. Dalalyan
Accepted, Stochastic Processes and their Applications
[arXiv]
Spectral Clustering with Variance Information for Group Structure Estimation in Panel Data
L. Yu, J. Gu, and S. Volgushev
Journal of Econometrics, 241(1), 2024, 105709.
Langevin Monte Carlo for Strongly Log-concave Distributions: Randomized Midpoint Revisited
L. Yu, A. Karagulyan, and A.S. Dalalyan
International Conference on Learning Representations (ICLR), 2024
[arXiv]
Mirror Descent Strikes Again: Optimal Stochastic Convex Optimization under Infinite Noise Variance
N.M. Vural, L. Yu, K. Balasubramanian, S. Volgushev, and M.A. Erdogdu
Conference on Learning Theory (COLT), 2022
An Analysis of Constant Step Size SGD in the Non-convex Regime: Asymptotic Normality and Bias
L. Yu, K. Balasubramanian, S. Volgushev, and M.A. Erdogdu
Advances in Neural Information Processing Systems (NeurIPS), 2021
False Discovery Rates in Biological Networks
L. Yu, T. Kaufmann, and J. Lederer
International Conference on Artificial Intelligence and Statistics (AISTATS), PMLR 2021, pages 163-171.
Oracle Inequalities for High-dimensional Prediction
J. Lederer, L. Yu, and I. Gayanova
Bernoulli, 25(2), 2019, pages 1225–1255.
Thesis:
Latent Structure Estimation for Panel Data and Theoretical Guarantees for Stochastic Optimization
Ph.D. thesis, 2022