Profiles2020-2026
Aryan Mokhtari
Department of Electrical and Computer Engineering, The University of Texas at Austin
27Papers
12Last author
24Unique co-authors
0First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers27
- Improving Online-to-Nonconvex Conversion for Smooth Optimization via Double OptimismAuthor 3 of 3ICLR 2026 · 3 citations
- Machine Unlearning under OverparameterizationAuthor 3 of 4NeurIPS 2025 · 4 citations
- On the Complexity of Finding Stationary Points in Nonconvex Simple Bilevel OptimizationAuthor 4 of 4NeurIPS 2025 · 3 citations
- Affine-Invariant Global Non-Asymptotic Convergence Analysis of BFGS under Self-ConcordanceAuthor 2 of 2NeurIPS 2025 · 2 citations
- Improved Complexity for Smooth Nonconvex Optimization: A Two-Level Online Learning Approach with Quasi-Newton MethodsAuthor 2 of 3STOC 2025 · 2 citations
- Provable Meta-Learning with Low-Rank AdaptationsAuthor 5 of 6NeurIPS 2025 · 2 citations
- Learning Mixtures of Experts with EM: A Mirror Descent PerspectiveAuthor 2 of 3ICML 2025
- In-Context Learning with Transformers: Softmax Attention Adapts to Function LipschitznessAuthor 3 of 5NeurIPS 2024 · 33 citations
- An Accelerated Gradient Method for Convex Smooth Simple Bilevel OptimizationAuthor 4 of 4NeurIPS 2024 · 16 citations
- Non-asymptotic Global Convergence Analysis of BFGS with the Armijo-Wolfe Line SearchAuthor 3 of 3NeurIPS 2024 · 15 citations
- Provable Multi-Task Representation Learning by Two-Layer ReLU Neural NetworksAuthor 4 of 5ICML 2024 · 15 citations
- Adaptive and Optimal Second-order Optimistic Methods for Minimax OptimizationAuthor 4 of 4NeurIPS 2024 · 12 citations
- Stochastic Newton Proximal Extragradient MethodAuthor 3 of 3NeurIPS 2024 · 2 citations
- Projection-Free Methods for Stochastic Simple Bilevel Optimization with Convex Lower-level ProblemAuthor 5 of 5NeurIPS 2023 · 21 citations
- Accelerated Quasi-Newton Proximal Extragradient: Faster Rate for Smooth Convex OptimizationAuthor 2 of 2NeurIPS 2023 · 14 citations
- Network Adaptive Federated Learning: Congestion and Lossy CompressionAuthor 3 of 3INFOCOM 2023 · 10 citations
- Greedy Pruning with Group Lasso Provably Generalizes for Matrix SensingAuthor 2 of 3NeurIPS 2023 · 4 citations
- FedAvg with Fine Tuning: Local Updates Lead to Representation LearningAuthor 3 of 4NeurIPS 2022 · 154 citations
- MAML and ANIL Provably Learn RepresentationsAuthor 2 of 4ICML 2022 · 38 citations
- Sharpened Quasi-Newton Methods: Faster Superlinear Rate and Larger Local Convergence NeighborhoodAuthor 4 of 4ICML 2022 · 17 citations
- Exploiting Shared Representations for Personalized Federated LearningAuthor 3 of 4ICML 2021 · 1,081 citations
- Generalization of Model-Agnostic Meta-Learning Algorithms: Recurring and Unseen TasksAuthor 2 of 3NeurIPS 2021 · 63 citations
- On the Convergence Theory of Debiased Model-Agnostic Meta-Reinforcement LearningAuthor 3 of 4NeurIPS 2021 · 31 citations
- Exploiting Local Convergence of Quasi-Newton Methods Globally: Adaptive Sample Size ApproachAuthor 2 of 2NeurIPS 2021 · 5 citations
- Personalized Federated Learning with Theoretical Guarantees: A Model-Agnostic Meta-Learning ApproachAuthor 2 of 3NeurIPS 2020 · 1,354 citations
- Task-Robust Model-Agnostic Meta-LearningAuthor 2 of 3NeurIPS 2020 · 66 citations
- Submodular Meta-LearningAuthor 2 of 3NeurIPS 2020
