Mladen Kolar
Professor of Data Sciences and Operations
University of Southern California Marshall School of Business
·
Mohamed bin Zayed University of Artificial Intelligence
·
Visiting Professor of Statistics and Data Science
Mladen Kolar is a Professor of Data Sciences and Operations at the University of Southern California Marshall School of Business and a Visiting Professor of Statistics and Data Science at Mohamed bin Zayed University of Artificial Intelligence. Before joining USC, he was on the faculty of the University of Chicago Booth School of Business. His research is focused on high-dimensional statistical methods, graphical models, varying-coefficient models and data mining, driven by the need to uncover interesting and scientifically meaningful structures from observational data. He is a Fellow of the Institute of Mathematical Statistics.
Interests
- Statistical machine learning
- Probabilistic graphical models
- Dynamic networks estimation
- High-dimensional estimation and inference
- Stochastic optimization with constraints
- Distributed optimization and federated learning
Education
- PhD in Machine Learning, 2013Carnegie Mellon University
- Diploma in Computer Engineering, 2006University of Zagreb, Faculty of Electrical Engineering and Computing
Publications
Trans-Glasso: A Transfer Learning Approach to Precision Matrix Estimation
Journal of the American Statistical Association
Simultaneous Inference for Covariance and Precision Matrices of Long-Range Dependent Time Series
IEEE Transactions on Information Theory, 72(6), 4246–4296
Provable Accelerated Bayesian Optimization with Knowledge Transfer
International Conference on Artificial Intelligence and Statistics (AISTATS)
Muon with Nesterov Momentum: Heavy-Tailed Noise and (Randomized) Inexact Polar Decomposition
Technical report
Gradient Clipping Beyond Vector Norms: A Spectral Approach for Matrix-Valued Parameters
Technical report
Fast and Consistent Structure Learning in Graphical Models via Approximate Cross-Validation
Advances in Neural Information Processing Systems (NeurIPS)
Confidence Sets for Causal Orderings
Journal of the American Statistical Association, 121(553), 690–703
AdAdaGrad: Adaptive Batch Size Schemes for Adaptive Gradient Methods
Statistical Learning and Data Science
Statistical Inference for Networks of High-Dimensional Point Processes
Journal of the American Statistical Association, 120(550), 1014–1024
High-Dimensional Differential Parameter Inference in Exponential Family using Time Score Matching
International Conference on Artificial Intelligence and Statistics (AISTATS)
Adaptive Client Sampling in Federated Learning via Online Learning with Bandit Feedback
Journal of Machine Learning Research, 26(8), 1–67
Adaptive Batch Size Schedules for Distributed Training of Language Models with Data and Model Parallelism
Conference on Parsimony and Learning (CPAL)
Trust-Region Sequential Quadratic Programming for Stochastic Optimization with Random Models
Technical report
Pessimism Meets Risk: Risk-Sensitive Offline Reinforcement Learning
International Conference on Machine Learning (ICML)
Latent Multimodal Functional Graphical Model Estimation
Journal of the American Statistical Association, 119(547), 2217–2229
Instrumental Variable Value Iteration for Causal Offline Reinforcement Learning
Journal of Machine Learning Research, 25(303), 1–56
Inconsistency of cross-validation for structure learning in Gaussian graphical models
International Conference on Artificial Intelligence and Statistics (AISTATS)
High-dimensional Functional Graphical Model Structure Learning via Neighborhood Selection Approach
Electronic Journal of Statistics, 18(1), 1042–1129
Fully Stochastic Trust-Region Sequential Quadratic Programming for Equality-Constrained Optimization Problems
SIAM Journal on Optimization, 34(2), 2007-2037
Convergence Analysis of Accelerated Stochastic Gradient Descent under the Growth Condition
Mathematics of Operations Research, 49(4), 2492–2526
Communication-Efficient Adaptive Batch Size Strategies for Distributed Local Gradient Methods
Technical report
A Fast Temporal Decomposition Procedure for Long-horizon Nonlinear Dynamic Programming
Mathematics of Operations Research, 49(2), 1012–1044
One Policy is Enough: Parallel Exploration with a Single Policy is Minimax Optimal for Reward-Free Reinforcement Learning
International Conference on Artificial Intelligence and Statistics (AISTATS)
Personalized Federated Learning: A Unified Framework and Universal Optimization Techniques
Transactions on Machine Learning Research
Inequality Constrained Stochastic Nonlinear Optimization via Active-Set Sequential Quadratic Programming
Mathematical Programming, 202(1), 279–353
Differentially Private Matrix Completion through Low-rank Matrix Factorization
International Conference on Artificial Intelligence and Statistics (AISTATS)
Constrained Optimization via Exact Augmented Lagrangian and Randomized Iterative Sketching
International Conference on Machine Learning (ICML)
Addressing Budget Allocation and Revenue Allocation in Data Market Environments Using an Adaptive Sampling Algorithm
International Conference on Machine Learning (ICML)
Provably training overparameterized neural network classifiers with non-convex constraints
Electronic Journal of Statistics, 16(2)
Gradient-Variation Bound for Online Convex Optimization with Constraints
AAAI Conference on Artificial Intelligence (to appear)
Pessimism meets VCG: Learning Dynamic Mechanism Design via Offline Reinforcement Learning
Proceedings of the 39th International Conference on Machine Learning (ICML) 2022
Dynamic Regret Minimization for Control of Non-stationary Linear Dynamical Systems
Proceedings of the 2022 ACM SIGMETRICS
An adaptive stochastic sequential quadratic programming with differentiable exact augmented lagrangians
Mathematical Programming
FuDGE: A Method to Estimate a Functional Differential Graph in a High-Dimensional Setting
Journal of Machine Learning Research
A Nonconvex Framework for Structured Dynamic Covariance Recovery
Journal of Machine Learning Research
Inference for high-dimensional varying-coefficient quantile regression
Electronic Journal of Statistics, 15(2)
Two-sample inference for high-dimensional Markov networks
Journal of the Royal Statistical Society. Series B. Statistical Methodology, 83(5)
Robust Inference for High-Dimensional Linear Models via Residual Randomization
Proceedings of the 38th International Conference on Machine Learning (ICML) 2021
Estimating differential latent variable graphical models with applications to brain connectivity
Biometrika, 108(2)
Tensor Canonical Correlation Analysis With Convergence and Statistical Guarantees
Journal of Computational and Graphical Statistics, 30(3)
Provably Efficient Neural Estimation of Structural Equation Model: An Adversarial Approach
Advances in Neural Information Processing Systems 33 (NeurIPS), 2020
Estimation of a Low-rank Topic-Based Model for Information Cascades
Journal of Machine Learning Research
Semiparametric Nonlinear Bipartite Graph Representation Learning with Provable Guarantees
Proceedings of the 37th International Conference on Machine Learning (ICML) 2020
Recovery of simultaneous low rank and two-way sparse coefficient matrices, a nonconvex approach
Electronic Journal of Statistics, 14(1)
Simultaneous Inference for Pairwise Graphical Models with Generalized Score Matching
Journal of Machine Learning Research
Kernel meets sieve: post-regularization confidence bands for sparse additive model
Journal of the American Statistical Association, 115(532)
Partially Linear Additive Gaussian Graphical Models
Proceedings of the 36th International Conference on Machine Learning
Learning Influence-Receptivity Network Structure with Guarantee
Proceedings of the 22nd International Conference on Artificial Intelligence and Statistics
High-dimensional Varying Index Coefficient Models via Stein's Identity
Journal of Machine Learning Research
Direct Estimation of Differential Functional Graphical Models
Advances in Neural Information Processing Systems, 32
Convergent Policy Optimization for Safe Reinforcement Learning
Advances in Neural Information Processing Systems, 32
ROCKET: Robust confidence intervals via Kendall's tau for transelliptical graphical models
Ann. Statist
Provable Gaussian Embedding with One Observation
Advances in Neural Information Processing Systems, 31
Post-Regularization Inference for Time-Varying Nonparanormal Graphical Models
Journal of Machine Learning Research
Joint Nonparametric Precision Matrix Estimation with Confounding
Uncertainty in Artificial Intelligence
Efficient Distributed Learning with Sparsity
Proceedings of the 34th International Conference on Machine Learning
Sketching Meets Random Projection in the Dual: A Provable Recovery Algorithm for Big and High-dimensional Data
Proceedings of the 20th International Conference on Artificial Intelligence and Statistics
Uniform inference for high-dimensional quantile regression: linear functionals and regression rank scores
arXiv preprint arXiv:1702.06209
The Expxorcist: Nonparametric Graphical Models Via Conditional Exponential Densities
Advances in Neural Information Processing Systems, 30
Sketching meets random projection in the dual: a provable recovery algorithm for big and high-dimensional data
Electron. J. Stat
Sketching meets random projection in the dual: a provable recovery algorithm for big and high-dimensional data
Electronic Journal of Statistics, 11(2), 4896–4944
An Influence-Receptivity Model for Topic based Information Cascades
2017 IEEE International Conference on Data Mining
Distributed Multi-Task Learning
Proceedings of the 19th International Conference on Artificial Intelligence and Statistics
Statistical Inference for Pairwise Graphical Models Using Score Matching
Advances in Neural Information Processing Systems, 29
Inference for High-dimensional Exponential Family Graphical Models
Proceedings of the 19th International Conference on Artificial Intelligence and Statistics
Learning structured densities via infinite dimensional exponential families
Advances in Neural Information Processing Systems, 28
A General Framework for Robust Testing and Confidence Regions in High-Dimensional Quantile Regression
ArXiv e-prints, arXiv:1412.8724
Mean and variance estimation in high-dimensional heteroscedastic models with non-convex penalties
ArXiv e-prints, arXiv:1410.7874
Markov Network Estimation From Multi-attribute Data
Proceedings of the 30th International Conference on Machine Learning
Feature Selection in High-Dimensional Classification
Proceedings of the 30th International Conference on Machine Learning
Variance Function Estimation in High-dimensions
Proceedings of the 29th International Conference on Machine Learning
Consistent Covariance Selection From Data With Missing Values
Proceedings of the 29th International Conference on Machine Learning
Marginal Regression For Multitask Learning
Proceedings of the 15th International Conference on Artificial Intelligence and Statistics
Statistical and computational tradeoffs in biclustering
NIPS 2011 Workshop on Computational Trade-offs in Statistical Learning
On Time Varying Undirected Graphs
Proceedings of the 14th International Conference on Artificial Intelligence and Statistics
Minimax Localization of Structural Information in Large Noisy Matrices
Advances in Neural Information Processing Systems, 24
Ultra-high Dimensional Multiple Output Learning With Simultaneous Orthogonal Matching Pursuit: Screening Approach
Proc. of AISTATS
CSMET: Comparative Genomic Motif Detection via Multi-Resolution Phylogenetic Shadowing
PLoS Computational Biology