Improved Complexity for Smooth Nonconvex Optimization: A Two-Level Online Learning Approach with Quasi-Newton Methods

Authors: Jiang, Ruichen, Mokhtari, Aryan, Patitucci, Francisco

Subject Terms: Mathematics - Optimization and Control, Computer Science - Machine Learning, Statistics - Machine Learning

Access URL: http://arxiv.org/abs/2412.02175

Learning Mixtures of Experts with EM

Authors: Fruytier, Quentin, Mokhtari, Aryan, Sanghavi, Sujay

Subject Terms: Computer Science - Machine Learning, Statistics - Machine Learning

Access URL: http://arxiv.org/abs/2411.06056

Meta-Learning Adaptable Foundation Models

Authors: Block, Jacob L., Srinivasan, Sundararajan, Collins, Liam, Mokhtari, Aryan, Shakkottai, Sanjay

Subject Terms: Computer Science - Machine Learning

Access URL: http://arxiv.org/abs/2410.22264

On the Crucial Role of Initialization for Matrix Factorization

Authors: Li, Bingcong, Zhang, Liang, Mokhtari, Aryan, He, Niao

Subject Terms: Computer Science - Machine Learning, Electrical Engineering and Systems Science - Signal Processing, Mathematics - Optimization and Control

Access URL: http://arxiv.org/abs/2410.18965

Online Learning Guided Quasi-Newton Methods with Global Non-Asymptotic Convergence

Authors: Jiang, Ruichen, Mokhtari, Aryan

Subject Terms: Mathematics - Optimization and Control, Computer Science - Machine Learning, Statistics - Machine Learning

Access URL: http://arxiv.org/abs/2410.02626

Convergence Analysis of Adaptive Gradient Methods under Refined Smoothness and Noise Assumptions

Authors: Jiang, Ruichen, Maladkar, Devyani, Mokhtari, Aryan

Subject Terms: Mathematics - Optimization and Control, Computer Science - Machine Learning, Statistics - Machine Learning

Access URL: http://arxiv.org/abs/2406.04592

Adaptive and Optimal Second-order Optimistic Methods for Minimax Optimization

Authors: Jiang, Ruichen, Kavis, Ali, Jin, Qiujiang, Sanghavi, Sujay, Mokhtari, Aryan

Subject Terms: Mathematics - Optimization and Control, Computer Science - Machine Learning, Statistics - Machine Learning

Access URL: http://arxiv.org/abs/2406.02016

Stochastic Newton Proximal Extragradient Method

Authors: Jiang, Ruichen, Dereziński, Michał, Mokhtari, Aryan

Subject Terms: Mathematics - Optimization and Control, Computer Science - Machine Learning, Statistics - Machine Learning

Access URL: http://arxiv.org/abs/2406.01478

Non-asymptotic Global Convergence Analysis of BFGS with the Armijo-Wolfe Line Search

Authors: Jin, Qiujiang, Jiang, Ruichen, Mokhtari, Aryan

Subject Terms: Mathematics - Optimization and Control

Access URL: http://arxiv.org/abs/2404.16731

Non-asymptotic Global Convergence Rates of BFGS with Exact Line Search

Authors: Jin, Qiujiang, Jiang, Ruichen, Mokhtari, Aryan

Subject Terms: Mathematics - Optimization and Control

Access URL: http://arxiv.org/abs/2404.01267

Meta-Learning for Image-Guided Millimeter-Wave Beam Selection in Unseen Environments

Authors: Gu, Jerry, Collins, Liam, Roy, Debashri, Mokhtari, Aryan, Shakkottai, Sanjay, Chowdhury, Kaushik R.

Source: ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) Acoustics, Speech and Signal Processing (ICASSP), ICASSP 2023 - 2023 IEEE International Conference on. :1-5 Jun, 2023

Relation: ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Network Adaptive Federated Learning: Congestion and Lossy Compression

Authors: Hegde, Parikshit, de Veciana, Gustavo, Mokhtari, Aryan

Source: IEEE INFOCOM 2023 - IEEE Conference on Computer Communications Computer Communications, IEEE INFOCOM 2023 - IEEE Conference on. :1-10 May, 2023

Relation: IEEE INFOCOM 2023 - IEEE Conference on Computer Communications

In-Context Learning with Transformers: Softmax Attention Adapts to Function Lipschitzness

Authors: Collins, Liam, Parulekar, Advait, Mokhtari, Aryan, Sanghavi, Sujay, Shakkottai, Sanjay

Subject Terms: Computer Science - Machine Learning, Computer Science - Artificial Intelligence, Computer Science - Computation and Language

Access URL: http://arxiv.org/abs/2402.11639

An Accelerated Gradient Method for Convex Smooth Simple Bilevel Optimization

Authors: Cao, Jincheng, Jiang, Ruichen, Hamedani, Erfan Yazdandoost, Mokhtari, Aryan

Subject Terms: Mathematics - Optimization and Control, Computer Science - Machine Learning, Statistics - Machine Learning

Access URL: http://arxiv.org/abs/2402.08097

Krylov Cubic Regularized Newton: A Subspace Second-Order Method with Dimension-Free Convergence Rate

Authors: Jiang, Ruichen, Raman, Parameswaran, Sabach, Shoham, Mokhtari, Aryan, Hong, Mingyi, Cevher, Volkan

Subject Terms: Mathematics - Optimization and Control, Computer Science - Machine Learning, Statistics - Machine Learning

Access URL: http://arxiv.org/abs/2401.03058

Leveraging Synergies Between AI and Networking to Build Next Generation Edge Networks

Authors: Lin, Sen, Shi, Ming, Arora, Anish, Bassily, Raef, Bertino, Elisa, Caramanis, Constantine, Chowdhury, Kaushik, Ekici, Eylem, Eryilmaz, Atilla, Ioannidis, Stratis, Jiang, Nan, Joshi, Gauri, Kurose, Jim, Liang, Yingbin, Lin, Zhiqiang, Liu, Jia, Liu, Mingyan, Melodia, Tommaso, Mokhtari, Aryan, Nowak, Rob, Oh, Sewoong, Parthasarathy, Srini, Peng, Chunyi, Seferoglu, Hulya, Shroff, Ness, Shakkottai, Sanjay, Srinivasan, Kannan, Talwalkar, Ameet, Yener, Aylin, Ying, Lei

Source: 2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC) CIC Collaboration and Internet Computing (CIC), 2022 IEEE 8th International Conference on. :16-25 Dec, 2022

Relation: 2022 IEEE 8th International Conference on Collaboration and Internet Computing (CIC)

Projection-Free Methods for Stochastic Simple Bilevel Optimization with Convex Lower-level Problem

Authors: Cao, Jincheng, Jiang, Ruichen, Abolfazli, Nazanin, Hamedani, Erfan Yazdandoost, Mokhtari, Aryan

Subject Terms: Mathematics - Optimization and Control, Computer Science - Machine Learning, Statistics - Machine Learning

Access URL: http://arxiv.org/abs/2308.07536

Provable Multi-Task Representation Learning by Two-Layer ReLU Neural Networks

Authors: Collins, Liam, Hassani, Hamed, Soltanolkotabi, Mahdi, Mokhtari, Aryan, Shakkottai, Sanjay

Subject Terms: Computer Science - Machine Learning

Access URL: http://arxiv.org/abs/2307.06887

Adaptive Node Participation for Straggler-Resilient Federated Learning

Authors: Reisizadeh, Amirhossein, Tziotis, Isidoros, Hassani, Hamed, Mokhtari, Aryan, Pedarsani, Ramtin

Source: ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) Acoustics, Speech and Signal Processing (ICASSP), ICASSP 2022 - 2022 IEEE International Conference on. :8762-8766 May, 2022

Relation: ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Limited-Memory Greedy Quasi-Newton Method with Non-asymptotic Superlinear Convergence Rate

Authors: Gao, Zhan, Mokhtari, Aryan, Koppel, Alec

Subject Terms: Mathematics - Optimization and Control, Computer Science - Machine Learning, Statistics - Machine Learning

Access URL: http://arxiv.org/abs/2306.15444