Aryan Mokhtari
Associate Professor
Studies optimization for machine learning, including meta-learning, federated learning and second-order methods.
Aryan Mokhtari's lab at UT Austin works on learning theory & optimization and trustworthy & safe ML. 29 papers since 2023. 4 PhD students. 6 alumni. Applications close .
At a glance
- papers, 2023–2026
- 8/ year
- at T1 venues
- 0%
- PhD students
- 4
- MS students
- 0
- alumni listed
- 6
Papers by yearfrom the lab homepage, 2023–2026
- 4 papers2026 · T1 0 · collecting
- 10 papers2025 · T1 0
- 7 papers2024 · T1 0
- 8 papers2023 · T1 0
T1 venuesOther
Research focuspaper titles that use each topic's words
- Learning theory & optimization16 papers
- meta-learning1 paper
- federated learning1 paper
- Trustworthy & safe ML0 papers
- studies optimization0 papers
Paper list
From the lab homepage.
20264 papers
- Conference on Learning Theory (COLT)Adaptive Matrix Online Learning through Smoothing with Guarantees for Nonsmooth Nonconvex Optimization
- International Conference on Learning Representations (ICLR)Improving Online-to-Nonconvex Conversion for Smooth Optimization via Double Optimism
- Conference on Learning Theory (COLT)Statistical Learning from Attribution Sets
- Conference on Machine Learning (ICML)Temper-Then-Tilt: Principled Unlearning for Generative Models through Tempering and Classifier Guidance
202510 papers
- Neural Information Processing Systems (NeurIPS), .Affine-Invariant Global Non-Asymptotic Convergence Analysis of BFGS under Self-Concordance
- SIOPTGeneralized Optimistic Methods for Convex-Concave Saddle Point Problems
- Symposium on Theory of Computing (STOC), .Improved Complexity for Smooth Nonconvex Optimization: A Two-Level Online Learning Approach with Quasi-Newton Methods
- Conference on Machine Learning (ICML)Learning Mixtures of Experts with EM: A Mirror Descent Perspective
- Neural Information Processing Systems (NeurIPS), .Machine Unlearning under Overparameterization
- Mathematical Programming (MAPR)Non-asymptotic Global Convergence Rates of BFGS with Exact Line Search
- Neural Information Processing Systems (NeurIPS), .On the Complexity of Finding Stationary Points in Nonconvex Simple Bilevel Optimization
- Int. Conference on Learning Representations (ICLR), .On the Crucial Role of Initialization for Matrix Factorization
- Conference on Learning Theory (COLT), .Provable Complexity Improvement of AdaGrad over SGD: Upper and Lower Bounds in Stochastic Non-Convex Optimization
- Neural Information Processing Systems (NeurIPS), .Provable Meta-Learning with Low-Rank Adaptations
20247 papers
- Neural Information Processing Systems (NeurIPS), .Adaptive and Optimal Second-order Optimistic Methods for Minimax Optimization
- Neural Information Processing Systems (NeurIPS), .An Accelerated Gradient Method for Simple Bilevel Optimization with Convex Lower-level Problem
- Int. Conf. on Artificial Intelligence and Statistics (AISTATS), .Krylov Cubic Regularized Newton: A Subspace Second-Order Method with Dimension-Free Convergence Rate
- Neural Information Processing Systems (NeurIPS), .Non-asymptotic Global Convergence Analysis of BFGS with the Armijo-Wolfe Line Search
- Int. Conference on Machine Learning (ICML), .Provable Multi-Task Representation Learning by Two-Layer ReLU Neural Networks
- Transactions on Machine Learning Research (TMLR), .Statistical and Computational Complexities of BFGS Quasi-Newton Method for Generalized Linear Models
- Neural Information Processing Systems (NeurIPS), .Stochastic Newton Proximal Extragradient Method
20238 papers
- Int. Conf. on Artificial Intelligence and Statistics (AISTATS), .A Conditional Gradient-based Method for Simple Bilevel Optimization with Convex Lower-level Problem
- Neural Information Processing Systems (NeurIPS), .Accelerated Quasi-Newton Proximal Extragradient: Faster Rate for Smooth Convex Optimization
- Neural Information Processing Systems (NeurIPS), .Greedy Pruning with Group Lasso Provably Generalizes for Matrix Sensing
- Conference on Learning Theory (COLT), .InfoNCE Loss Provably Learns Cluster-Preserving Representations
- Int. Conf. on Computer Communications (INFOCOM)Network Adaptive Federated Learning: Congestion and Lossy Compression
- Conference on Learning Theory (COLT), .Online Learning Guided Curvature Approximation: A Quasi-Newton Method with Global Non-Asymptotic Superlinear Convergence
- Neural Information Processing Systems (NeurIPS), .Projection-Free Methods for Stochastic Simple Bilevel Optimization with Convex Lower-level Problem
- IEEE Transactions on Information TheoryProvably Private Distributed Averaging Consensus: An Information-Theoretic Approach
Where alumni went6 listed
Apply to UT AustinFall 2027
Computer Science PhD
- Round
- Single round (Dec 15)
- Opens
- GRE
- See program page
- English
- See program page
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- GBImperial
Yves-Alexandre de Montjoye
- Trustworthy & safe ML
- USNYU
Mehryar Mohri
- Learning theory & optimization
- USCMU
Maria-Florina Balcan
- Learning theory & optimization
- USUMD
Yizheng Chen
- Trustworthy & safe ML
- USMichigan
Clayton Scott
- Learning theory & optimization
- USUC Berkeley
Somayeh Sojoudi
- Trustworthy & safe ML
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
UT Austincloses