AM
Aryan Mokhtari
cs.LGmath.OCstat.MLcs.DCcs.AIcs.CLcs.DScs.ITmath.ITcs.MA
On Valency
published · living versionsW_un5mvf9d·v1 · currentpublished
Improved Gradient Complexity for Quasi-Newton Proximal Extragradient Methods via Certified Curvature Reuse
with Francisco Patitucci
1 version
Preprints & journals
105 papers in the corpus · 2014–2026Stochastic Dynamic Barrier Perturbed Gradient Methods for Nonconvex Simple Bilevel Optimization2607.10957v2 · Mohammad Mahdi Ahmadi, Jincheng Cao, Aryan Mokhtari et al.2026 · 0 citationsarXiv
Curriculum Learning-Guided Progressive Distillation in Large Language Models2605.11260v1 · Jincheng Cao, Fanzhi Zeng, Leqi Liu et al.2026 · 0 citationsarXiv
Aletheia tackles FirstProof autonomously2602.21201v3 · Tony Feng, Junehyuk Jung, Sang-hyun Kim et al.2026 · 0 citationsarXiv
Adaptive Optimization via Momentum on Variance-Normalized Gradients2602.10204v2 · Francisco Patitucci, Aryan Mokhtari2026 · 0 citationsarXiv
Temper-Then-Tilt: Principled Unlearning for Generative Models through Tempering and Classifier Guidance2602.10217v2 · Jacob L. Block, Mehryar Mohri, Aryan Mokhtari et al.2026 · 0 citationsarXiv
Adaptive Matrix Online Learning through Smoothing with Guarantees for Nonsmooth Nonconvex Optimization2602.08232v1 · Ruichen Jiang, Zakaria Mhammedi, Mehryar Mohri et al.2026 · 0 citationsarXiv
Statistical Learning from Attribution Sets2602.06276v2 · Lorne Applebaum, Robert Busa-Fekete, August Y. Chen et al.2026 · 0 citationsarXiv
CARE-RFT: Confidence-Anchored Reinforcement Finetuning for Reliable Reasoning in Large Language Models2602.00085v1 · Shuozhe Li, Jincheng Cao, Bodun Hu et al.2026 · 0 citationsarXiv
Sculpting the Latent Space: Disentanglement with WAE-MMD and Programmable Priors2510.11953v2 · Quentin Fruytier, Akshay Malhotra, Shahab Hamidi-Rad et al.2025 · 0 citationsarXiv
Improving Online-to-Nonconvex Conversion for Smooth Optimization via Double Optimism2510.03167v3 · Francisco Patitucci, Ruichen Jiang, Aryan Mokhtari2025 · 0 citationsarXiv
On the Complexity of Finding Stationary Points in Nonconvex Simple Bilevel Optimization2507.23155v1 · Jincheng Cao, Ruichen Jiang, Erfan Yazdandoost Hamedani et al.2025 · 0 citationsarXiv
Affine-Invariant Global Non-Asymptotic Convergence Analysis of BFGS under Self-Concordance2507.00361v2 · Qiujiang Jin, Aryan Mokhtari2025 · 0 citationsarXiv
Online Learning-guided Learning Rate Adaptation via Gradient Alignment2506.08419v1 · Ruichen Jiang, Ali Kavis, Aryan Mokhtari2025 · 0 citationsarXiv
Machine Unlearning under Overparameterization2505.22601v2 · Jacob L. Block, Aryan Mokhtari, Sanjay Shakkottai2025 · 0 citationsarXiv
Improved Complexity for Smooth Nonconvex Optimization: A Two-Level Online Learning Approach with Quasi-Newton Methods2412.02175v1 · Ruichen Jiang, Aryan Mokhtari, Francisco Patitucci2024 · 2 citationsarXiv
Learning Mixtures of Experts with EM: A Mirror Descent Perspective2411.06056v2 · Quentin Fruytier, Aryan Mokhtari, Sujay Sanghavi2024 · 0 citationsProceedings of the 42nd International Conference on Machine Learning, PMLR 267:17690-17726, 2025
Provable Meta-Learning with Low-Rank Adaptations2410.22264v2 · Jacob L. Block, Sundararajan Srinivasan, Liam Collins et al.2024 · 0 citationsarXiv
On the Crucial Role of Initialization for Matrix Factorization2410.18965v3 · Bingcong Li, Liang Zhang, Aryan Mokhtari et al.2024 · 0 citationsarXiv
Online Learning Guided Quasi-Newton Methods with Global Non-Asymptotic Convergence2410.02626v1 · Ruichen Jiang, Aryan Mokhtari2024 · 1 citationarXiv
Provable Multi-Task Representation Learning by Two-Layer ReLU Neural Networks.39421278 · Collins, Liam, Hassani, Hamed, Soltanolkotabi, Mahdi et al.2024 · 0 citationsProceedings of machine learning research. 2024;235:9292-9345
Provable Complexity Improvement of AdaGrad over SGD: Upper and Lower Bounds in Stochastic Non-Convex Optimization2406.04592v3 · Ruichen Jiang, Devyani Maladkar, Aryan Mokhtari2024 · 0 citationsarXiv
Adaptive and Optimal Second-order Optimistic Methods for Minimax Optimization2406.02016v2 · Ruichen Jiang, Ali Kavis, Qiujiang Jin et al.2024 · 1 citationarXiv
Stochastic Newton Proximal Extragradient Method2406.01478v2 · Ruichen Jiang, MichałDerezi'nski, Aryan Mokhtari2024 · 1 citationarXiv
Non-asymptotic Global Convergence Analysis of BFGS with the Armijo-Wolfe Line Search2404.16731v2 · Qiujiang Jin, Ruichen Jiang, Aryan Mokhtari2024 · 2 citationsarXiv
Non-asymptotic Global Convergence Rates of BFGS with Exact Line Search2404.01267v2 · Qiujiang Jin, Ruichen Jiang, Aryan Mokhtari2024 · 4 citationsarXiv
In-Context Learning with Transformers: Softmax Attention Adapts to Function Lipschitzness2402.11639v2 · Liam Collins, Advait Parulekar, Aryan Mokhtari et al.2024 · 2 citationsarXiv
An Accelerated Gradient Method for Convex Smooth Simple Bilevel Optimization2402.08097v2 · Jincheng Cao, Ruichen Jiang, Erfan Yazdandoost Hamedani et al.2024 · 1 citationarXiv
Krylov Cubic Regularized Newton: A Subspace Second-Order Method with Dimension-Free Convergence Rate2401.03058v1 · Ruichen Jiang, Parameswaran Raman, Shoham Sabach et al.2024 · 1 citationarXiv
Projection-Free Methods for Stochastic Simple Bilevel Optimization with Convex Lower-level Problem2308.07536v1 · Jincheng Cao, Ruichen Jiang, Nazanin Abolfazli et al.2023 · 1 citationarXiv
Limited-Memory Greedy Quasi-Newton Method with Non-asymptotic Superlinear Convergence Rate2306.15444v2 · Zhan Gao, Aryan Mokhtari, Alec Koppel2023 · 0 citationsarXiv
An Inexact Conditional Gradient Method for Constrained Bilevel Optimization2306.02429v2 · Nazanin Abolfazli, Ruichen Jiang, Aryan Mokhtari et al.2023 · 1 citationarXiv
Accelerated Quasi-Newton Proximal Extragradient: Faster Rate for Smooth Convex Optimization2306.02212v1 · Ruichen Jiang, Aryan Mokhtari2023 · 0 citationsarXiv
Greedy Pruning with Group Lasso Provably Generalizes for Matrix Sensing2303.11453v2 · Nived Rajaraman, Devvrit, Aryan Mokhtari et al.2023 · 1 citationarXiv
Online Learning Guided Curvature Approximation: A Quasi-Newton Method with Global Non-Asymptotic Superlinear Convergence2302.08580v2 · Ruichen Jiang, Qiujiang Jin, Aryan Mokhtari2023 · 1 citationarXiv
InfoNCE Loss Provably Learns Cluster-Preserving Representations2302.07920v1 · Advait Parulekar, Liam Collins, Karthikeyan Shanmugam et al.2023 · 14 citationsarXiv
Network Adaptive Federated Learning: Congestion and Lossy Compression2301.04430v1 · Parikshit Hegde, Gustavo de Veciana, Aryan Mokhtari2023 · 11 citationsarXiv
Conditional Gradient Methods2211.14103v5 · G'abor Braun, Alejandro Carderera, Cyrille W. Combettes et al.2022 · 14 citationsarXiv
Future Gradient Descent for Adapting the Temporal Shifting Data Distribution in Online Recommendation Systems2209.01143v1 · Mao Ye, Ruichen Jiang, Haoxiang Wang et al.2022 · 2 citationsarXiv
A Conditional Gradient-based Method for Simple Bilevel Optimization with Convex Lower-level Problem2206.08868v3 · Ruichen Jiang, Nazanin Abolfazli, Aryan Mokhtari et al.2022 · 4 citationsarXiv
Straggler-Resilient Personalized Federated Learning2206.02078v1 · Isidoros Tziotis, Zebang Shen, Ramtin Pedarsani et al.2022 · 4 citationsarXiv
Statistical and Computational Complexities of BFGS Quasi-Newton Method for Generalized Linear Models2206.00207v2 · Qiujiang Jin, Tongzheng Ren, Nhat Ho et al.2022 · 2 citationsarXiv
FedAvg with Fine Tuning: Local Updates Lead to Representation Learning2205.13692v1 · Liam Collins, Hamed Hassani, Aryan Mokhtari et al.2022 · 87 citationsarXiv
Sharpened Quasi-Newton Methods: Faster Superlinear Rate and Larger Local Convergence Neighborhood2202.10538v2 · Qiujiang Jin, Alec Koppel, Ketan Rajawat et al.2022 · 1 citationarXiv
Generalized Optimistic Methods for Convex-Concave Saddle Point Problems2202.09674v2 · Ruichen Jiang, Aryan Mokhtari2022 · 4 citationsarXiv
Provably Private Distributed Averaging Consensus: An Information-Theoretic Approach2202.09398v1 · Mohammad Fereydounian, Aryan Mokhtari, Ramtin Pedarsani et al.2022 · 6 citationsarXiv
The Power of Adaptivity in SGD: Self-Tuning Step Sizes with Unbounded Gradients and Affine Variance2202.05791v2 · Matthew Faw, Isidoros Tziotis, Constantine Caramanis et al.2022 · 5 citationsarXiv
MAML and ANIL Provably Learn Representations2202.03483v2 · Liam Collins, Aryan Mokhtari, Sewoong Oh et al.2022 · 10 citationsarXiv
Minimax Optimization: The Case of Convex-Submodular2111.01262v1 · Arman Adibi, Aryan Mokhtari, Hamed Hassani2021 · 1 citationarXiv
Exploiting Local Convergence of Quasi-Newton Methods Globally: Adaptive Sample Size Approach2106.05445v2 · Qiujiang Jin, Aryan Mokhtari2021 · 1 citationarXiv
Exploiting Shared Representations for Personalized Federated Learning2102.07078v3 · Liam Collins, Hamed Hassani, Aryan Mokhtari et al.2021 · 453 citationsarXiv
Career total: 167 works. 105 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.