World's Best Scientists 2026 revealed!

Overview

Mahdi Soltanolkotabi is affiliated with the University of Southern California in the United States. Their research primarily lies within the field of Computer Science, with a strong focus on subfields such as Artificial Intelligence, Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging, Computational Mechanics, and Statistics and Probability.

Their work spans several key topics including Domain Adaptation and Few-Shot Learning, Machine Learning and Algorithms, Sparse and Compressive Sensing Techniques, Privacy-Preserving Technologies in Data, Stochastic Gradient Optimization Techniques, Advanced Neural Network Applications, and Statistical Methods and Inference.

They have published extensively, with 42 papers in arXiv (Cornell University), 5 in PubMed, 2 in Information and Inference A Journal of the IMA, 1 in IEEE Journal on Selected Areas in Information Theory, and 1 in Findings of the Association for Computational Linguistics: NAACL 2022.

Recent papers include:

  • A Field Guide to Federated Optimization, 2021, arXiv (Cornell University)
  • Toward Moderate Overparameterization: Global Convergence Guarantees for Training Shallow Neural Networks, 2020, IEEE Journal on Selected Areas in Information Theory
  • FedNLP: Benchmarking Federated Learning Methods for Natural Language Processing Tasks, 2022, Findings of the Association for Computational Linguistics: NAACL 2022
  • On the Linear Convergence of Random Search for Discrete-Time LQR, 2020, IEEE Control Systems Letters
  • HUMUS-Net: Hybrid unrolled multi-scale network architecture for accelerated MRI reconstruction, 2022, arXiv (Cornell University)

Frequent co-authors in their work include Zalan Fabian, Salman Avestimehr, Chaoyang He, Samet Oymak, and Reinhard Heckel.

Best Publications

  • Phase Retrieval via Wirtinger Flow: Theory and Algorithms

    Emmanuel J. Candes;Xiaodong Li;Mahdi Soltanolkotabi

  • A Geometric Analysis of Subspace Clustering with Outliers

    Mahdi Soltanolkotabi;Emmanuel J. Candés

  • Phase retrieval from coded diffraction patterns

    Emmanuel J. Candès;Xiaodong Li;Mahdi Soltanolkotabi

  • Robust subspace clustering

    Mahdi Soltanolkotabi;Ehsan Elhamifar;Emmanuel J. Candès

  • Theoretical Insights Into the Optimization Landscape of Over-Parameterized Shallow Neural Networks

    Mahdi Soltanolkotabi;Adel Javanmard;Jason D. Lee

  • Low-rank solutions of linear matrix equations via procrustes flow

    Stephen Tu;Ross Boczar;Max Simchowitz;Mahdi Soltanolkotabi

  • Experimental robustness of Fourier ptychography phase retrieval algorithms

    Li-Hao Yeh;Jonathan Dong;Jingshan Zhong;Lei Tian

  • Compressed Sensing with Deep Image Prior and Learned Regularization

    Dave Van Veen;Ajil Jalal;Mahdi Soltanolkotabi;Eric Price

  • LAGRANGE CODED COMPUTING: OPTIMAL DESIGN FOR RESILIENCY, SECURITY, AND PRIVACY

    Salman Avestimehr;Mohammadreza Mousavi Kalan;Netanel Raviv;Mahdi Soltanolkotabi

  • Toward Moderate Overparameterization: Global Convergence Guarantees for Training Shallow Neural Networks

    Samet Oymak;Mahdi Soltanolkotabi

  • Gradient Descent with Early Stopping is Provably Robust to Label Noise for Overparameterized Neural Networks

    Mingchen Li;Mahdi Soltanolkotabi;Samet Oymak

  • A Field Guide to Federated Optimization

    Jianyu Wang;Zachary Charles;Zheng Xu;Gauri Joshi

  • A unified approach to sparse signal processing

    Farokh Marvasti;Arash Amini;Farzan Haddadi;Mehdi Soltanolkotabi

  • Lagrange Coded Computing: Optimal Design for Resiliency, Security and Privacy

    Qian Yu;Songze Li;Netanel Raviv;Seyed Mohammadreza Mousavi Kalan

  • Sharp Time–Data Tradeoffs for Linear Inverse Problems

    Samet Oymak;Benjamin Recht;Mahdi Soltanolkotabi

  • Learning ReLUs via Gradient Descent

    Mahdi Soltanolkotabi

  • Overparameterized Nonlinear Learning: Gradient Descent Takes the Shortest Path?

    Samet Oymak;Mahdi Soltanolkotabi

  • Towards moderate overparameterization: global convergence guarantees for training shallow neural networks

    Samet Oymak;Mahdi Soltanolkotabi

  • Convergence and sample complexity of gradient methods for the model-free linear quadratic regulator problem

    Hesameddin Mohammadi;Armin Zare;Mahdi Soltanolkotabi;Mihailo R. Jovanovic

  • Gradient Methods for Submodular Maximization

    S. Hamed Hassani;Mahdi Soltanolkotabi;Amin Karbasi

Frequent Co-Authors

A. Salman Avestimehr
A. Salman Avestimehr University of Southern California
Emmanuel J. Candès
Emmanuel J. Candès Stanford University
Farokh Marvasti
Farokh Marvasti Sharif University of Technology
Benjamin Recht
Benjamin Recht University of California, Berkeley
Ilias Diakonikolas
Ilias Diakonikolas University of Wisconsin–Madison
Richard M. Leahy
Richard M. Leahy University of Southern California
Rong Ge
Rong Ge Duke University
Justin P. Haldar
Justin P. Haldar University of Southern California
Jason D. Lee
Jason D. Lee Princeton University
Shankar Sastry
Shankar Sastry University of California, Berkeley

If you think any of the details on this page are incorrect, let us know.

Report an issue

We appreciate your kind effort to assist us to improve this page, it would be helpful providing us with as much detail as possible in the text box below:

Related Online Degrees & Career Pathways

Exploring Computer Science in the USA opens doors to a wide range of online degrees and flexible career options. For students looking to diversify, online electrical engineering career outcomes demonstrate the growing demand for tech-driven roles in engineering and technology sectors.

Not everyone needs a traditional four-year degree to advance. Shorter certificate programs that pay well are increasingly popular, offering practical skills and a fast track to employment in fields like IT support, networking, and cybersecurity.

If you're considering postgraduate education, the fastest masters degree online options can help you earn an advanced qualification quickly, often in just one year. This flexibility lets you balance work and study while advancing your career.

For those seeking long-term growth and job security, reviewing the best masters degree to get provides valuable insight into degrees most valued by employers, such as data science, artificial intelligence, and information systems.

Best Scientists Citing Mahdi Soltanolkotabi

Trending Scientists

Recently Published Articles