Associate Professor
Andrew and Erna Viterbi Early Career Chair
Departments of Industrial and Systems Engineering, Electrical and Computer Engineering, Quantitative and Computational Biology, and Computer Science
Associate Director of the USC-Meta Center for Research and Education in AI and Learning
USC Center for Systems and Control
Email: razaviya @ usc . edu
Google Scholar Page
Curriculum Vitae
Biography
Meisam Razaviyayn is an associate professor of Industrial and Systems Engineering, Computer Science, Quantitative and Computational Biology, and Electrical Engineering at the University of Southern California. He is also the associate director of the USC- Meta Center for Research and Education in AI and Learning; and also is a faculty visitor at Google Research. Prior to joining USC, he was a postdoctoral research fellow in the Department of Electrical Engineering at Stanford University and was a Visiting Scientist at the Simons Institute for the Theory of Computing at UC, Berkeley. He received his Ph.D. in Electrical Engineering with a minor in Computer Science at the University of Minnesota. He obtained his M.Sc. degree in Mathematics from the University of Minnesota. Meisam Razaviyayn is the recipient of the 2022 NSF CAREER Award, the 2022 Northrop Grumman Excellence in Teaching Award, the 2021 AFOSR Young Investigator Award, the 2021 3M Nontenured Faculty Award, 2020 ICCM Best Paper Award in Mathematics, IEEE Data Science Workshop Best Paper Award in 2019, the Signal Processing Society Young Author Best Paper Award in 2014, and the finalist for Best Paper Prize for Young Researcher in Continuous Optimization in 2013 and 2016. He is also the silver medalist of Iran’s National Mathematics Olympiad. His research interests include the design and the study of the fundamental aspects of optimization algorithms that arise in the modern data science era.
Recent News
- Our project “Making Generative AI Accessible: Efficient Training and Fine-Tuning with Resource-Aware Algorithms” is Funded by the USC Office of Research and Innovation, Generative AI Research Program, December 2024
- Our group received a gift funding from Amazon to improve Differentially Private training of large models, December 2024
- Our paper “Doppler: Differentially private optimizers with low-pass filter for privacy noise reduction” is published in NeurIPS 2024, December 2024
- Our paper “Incentive Systems for Fleets of New Mobility Services” is published in the IEEE Transactions on Intelligent Transportation Systems, December 2024
- Tianjian Huang defended his PhD, December 2024
- Ali Ghafelebashi defended his PhD, December 2024
- We receive funding from NSF to support students attending the International Conference on Continuous Optimization (ICCOPT), November 2024
- I joined the editorial board of the SIAM Journal On Optimization as an Associate Editor, November 2024
- I joined the editorial board of the Transactions on Machine Learning Research (TMLR) as an Action Editor, November 2024
- Our work “A Stochastic Optimization Framework for Private and Fair Learning From Decentralized Data” is available on arXiv, November 2024
- Devansh Gupta presented our work “On the Inherent Privacy of Two Point Zeroth Order Projected Gradient Descent” as an oral presentation in the Optimization for Machine Learning Workshop, October 2024
- Our work “Addax: Utilizing Zeroth-Order Gradients to Improve Memory Efficiency and Performance of SGD for Fine-Tuning Language Models” is available on arXiv, October 2024
- Our work “Adaptively Private Next-Token Prediction of Large Language Models” is available on arXiv, October 2024
- Our work “Disk: Differentially private optimizer with simplified Kalman filter for noise reduction” is available on arXiv, October 2024
- Our paper “Tradeoffs between convergence rate and noise amplification for momentum-based accelerated optimization algorithms” is published in the IEEE Transactions on Automatic Control, September 2024
- I joined Google Research as a Faculty Visitor, August 2024
- Mahdi Salmani joined our group! Welcome Mahdi! August 2024
- NSF NRT Program supports our project “NRT-AI: Integrating Artificial Intelligence and Op- erations Research Technologies.” We are creating the ORAI program, a PhD certificate program at the frontier of AI for decision-making, July 2024
- I will serve as the Area Chair in AISTATS 2024, June 2024
- Our paper “Differentially Private Next-Token Prediction of Large Language Models” is published in NAACL 2024, June 2024
- Sina Baharlouei won the Best Dissertation Award of the Department, April 2024
- Sina Baharlouei defended his PhD and will join eBay as an Applied Research Scientist, April 2024
- Our paper “Neural network-based score estimation in diffusion models: Optimization and generalization” is accepted in ICLR 2024, January 2024
- Our paper “f-FERM: A Scalable Framework for Robust Fair Empirical Risk Minimization” is accepted in ICLR 2024, January 2024
- Our project “A GPU Cluster for Empowering Research on Distributed Optimization and Learning in Massive-Scale Systems” is funded by AFOSR, November 2023
- I will serve the Workshop Chair of the Uncertainty in Artificial Intelligence (UAI) 2024 conference. I look forward to an exciting and fun conference. November 2023
- Our project “Private Learning With Public Data: From Theory to Practice and Back” is funded by Google Research, November 2023
- Ali Ghafelebashi won the Directors’ Award for Best Research Translation in Three Minute Thesis, November 2023
- ICCOPT 2025 Website is up! November 2023
- Invited Talk at EPFL CIS Colloquium on “A scalable stochastic optimization framework for robust and private fair learning”, October 2023
- Invited Talk at the ETH Zurich Universit on scalable fair learning, September 2023
- Our paper “RIFLE: Imputation and Robust Inference from Low Order Marginals” is accepted in Transactions on Machine Learning Research, September 2023
- Ali Ghafelebashi awarded ITS California and California Transportation Foundation Scholarship, August 2023
- Our group attended ICML 2023 and presented several posters
- Our paper “Optimal differentially private learning with public data” is available on arXiv, June 2023
- Our paper “Four Axiomatic Characterizations of the Integrated Gradients Attribution Method” is available on arXiv
- Andrew Lowy won the Center for the Applied Mathematical Sciences (CAMS) Graduate Student Prize, May 2023
- Our paper “Stochastic Differentially Private and Fair Learning” is presented in ICLR 2023, May 2023
- Our paper “Distributing Synergy Functions: Unifying Game-Theoretic Interaction Methods for Machine-Learning Explainability” is accepted in ICML 2023, May 2023
- Our paper “Improving Adversarial Robustness via Joint Classification and Multiple Explicit Detection Classes” is presented at AISTATS 2023, April 2023
- Our paper “Private non-convex federated learning without a trusted server” is presented in AISTATS 2023, April 2023
- Our paper “Congestion reduction via personalized incentives” is accepted by the Transportation Research Part C: Emerging Technologies, April 2023
- Presented our work “Fair and Private Backpropagation: A Scalable Framework for Fair and Private Learning” in the USC-Amazon Center on Secure & Trusted ML, April 2023
- I am honored to be invited by the National Academy of Engineering to attend the German-American Frontiers of Engineering Symposium, March 2023
- Hesameddin Mohammadi (advisor: Mihailo Jovanovic) won the Sun Prize (the first prize) of the Information Theory and Applications (ITA) graduation day talk presenting our joint work on Robustness of gradient-based methods for data-driven decision making, February 2023
- Our paper “Private Stochastic Optimization with Large Worst-Case Lipschitz Parameter: Optimal Rates for (Non-Smooth) Convex Losses and Extension to Non-Convex Losses” is presented at the International Conference on Algorithmic Learning Theory, February 2023.
- Presented our work “Stochastic Differentially Private and Fair Learning” in the Information Theory and Applications (ITA) Workshop, February 2023
- Our paper “I-CONVEX: Fast and Accurate de Novo Transcriptome Recovery from Long Reads” is accepted in the Machine Learning and Principles and Practice of Knowledge Discovery in Databases, January 2023
- Presented our research efforts on Responsible AI in Health and Medicine in Engineering for Mental Health Workshop at USC, January 2023
- Our work “A stochastic optimization framework for fair risk minimization” is published in Transactions on Machine Learning Research, December 2022.
- Invited Talk at the ML+AI Symposium @ USC, November 2022
- Our work “Tradeoffs between convergence rate and noise amplification for momentum-based accelerated optimization algorithms” is available on arXiv, September 2022
- Presented our work “Fair Backpropagation: A Scalable Framework for Fair Empirical Risk Minimization” in the USC-Amazon Center on Secure & Trusted ML, September 2022
- I am honored to be invited as a panelist discussing academic careers in the North American School of Information Theory, August 2022
- Presented our work at The International Conference on Continuous Optimization (ICCOPT), Lehigh University, July 2022
- I am honored to receive the Viterbi Northrop Grumman Excellence in Teaching Award, 2022
- Presented our research efforts on Responsible AI in Health and Medicine at Oracle (April 2022)
- I will be serving as the Associate Editor of the IEEE Transactions on Signal Processing (April 2022 – March 2024).
- Presented our group’s research on robust learning at the WPI Data Science Colloquium (April 2022)
- “Zeroth-Order Algorithms for Nonconvex-Strongly-Concave Minimax Problems with Improved Complexities“, has been accepted for publication in the Journal of Global Optimization (April 2022)
- A Rigorous Study of Integrated Gradients Method and Extensions to Internal Neuron Attributions is acctepted to ICML 2022. Joint work with Daniel Lundstrom and Tianjian Huang, April 2022.
- Received the NSF CAREER Award (March 2022)
- Presented some of our group’s research work on min-max optimization at the H. Milton Stewart School of Industrial & Systems Engineering, Georgia Institute of Technology (March 2022)
- Andrew Lowy and Dmitrii Ostrovskii presented their work at the INFORMS Optimization Society Conference (March 2022)
- Presented some of our group’s research work on min-max optimization in Simons Insitute Workshop on Adversarial Approaches in Machine Learning. Video of the talk. (Feb 2022)
- We launched the USC-Meta Center for Research and Education in AI and Learning (January 2022)
- Presented some of our group works in One World Optimization Seminar. YouTube video (January 2022)
- I am honored that my research proposal is selected for AFOSR Young Investigator Research Program (YIP) Award (November 2021).
- DAIR: Data augmented invariant regularization is available on arXiv. Joint work with Tianjian Huang, Chinnadhurai Sankar, Pooyan Amini, Satwik Kottur, Alborz Geramifard, and Ahmad Beirami (October 2021).
- Nonconvex-Nonconcave Min-Max Optimization with a Small Maximization Domain is available on arXiv. Joint work with Babak Barazandeh and Dmitrii Ostrovskii (October 2021).
- RIFLE is available on GitHub! RIFLE is a large-scale data imputation package designed based on a distributionally robust optimization framework. Check RIFLE package and the paper! (September 2021).
- Dmitrii Ostrovskii joined the Mathematics department at the University of Southern California as an Assistant Professor of Mathematics (August 2021).
- Tianjian Huang is doing an internship at Facebook AI (June-December 2021).
- Sina Baharlouei is doing an internship at the Bosch Center for Artificial Intelligence (June-October 2021).
- FERMI: Fair empirical risk minimization via exponential Rényi mutual information accepted to ICML-21 Workshop on Socially Responsible Machine Learning for oral presentation. Joint work with Andrew Lowy, Rakesh Pavan Sina Baharlouei, and Ahmad Beirami (July 2021).
- DAIR: Data augmented invariant regularization accepted to ICML-21 Workshop on Uncertainly and Robustness in Deep Learning. Joint work with Tianjian Huang, Chinnadhurai Sankar, Pooyan Amini, Satwik Kottur, Alborz Geramifard, and Ahmad Beirami (July 2021).
- Efficient Search of First-Order Nash Equilibria in Nonconvex-Concave Smooth Min-Max Problems accepted to SIAM Journal on Optimization. Joint work with Dmitrii M. Ostrovskii and Andrew Lowy (July 2021).
- Linearized ADMM Converges to Second-Order Stationary Points for Non-Convex Problems published in IEEE Transactions on Signal Processing. Joint work with Songtao Lu, Jason Lee, and Mingyi Hong (July 2021).
- Locally Differentially Private Federated Learning: Efficient Algorithms with Tight Risk Bounds is available on arXiv. Joint work with Andrew Lowy (June 2021).
- Efficient Algorithms for Estimating the Parameters of Mixed Linear Regression Models is available on arXiv. Joint work with Babak Barazandeh, Ali Ghafelebashi, and Ram Sriharsha (May 2021).
- Meisam Razaviyayn won the 3M’s Non-Tenured Faculty Award (3M NTFA), 2021 (May 2021).
- Ahmad Beirami, Emily Black, Krishna Gummadi, Hoda Heidari, Baharan Mirzasoleiman, Meisam Razaviyayn, and Joshua Williams co-organized the Responsible AI Workshop in ICLR (April 2021).
- Alternating direction method of multipliers for quantization” is published in AISTATS 2021. A joint work with Tianjian Huang, Prajwal Singhania, Maziar Sanjabi, and Pabitra Mitra (April 2021).
- Congestion Reduction via Personalized Incentives is available online. Joint work with Ali Ghafelebashi and Maged Dessouky (April 2021).
- I was honored to be invited as a panelist discussing academic careers in the North American School of Information Theory, August 2022
- Presented our work in The International Conference on Continuous Optimization (ICCOPT), Lehigh University, July 2022
- Presented our research efforts on Responsible AI in Health and Medicine at Oracle (April 2022)
- I will be serving as the Associate Editor of the IEEE Transactions on Signal Processing (April 2022 – March 2024).
- Presented our group’s research on robust learning at the WPI Data Science Colloquium (April 2022)
- “Zeroth-Order Algorithms for Nonconvex-Strongly-Concave Minimax Problems with Improved Complexities“, has been accepted for publication in the Journal of Global Optimization (April 2022)
- Received the NSF CAREER Award (March 2022)
- Presented some of our group’s research work on min-max optimization at the H. Milton Stewart School of Industrial & Systems Engineering, Georgia Institute of Technology (March 2022)
- Andrew Lowy and Dmitrii Ostrovskii presented their work at the INFORMS Optimization Society Conference (March 2022)
- Presented some of our group’s research work on min-max optimization in Simons Insitute Workshop on Adversarial Approaches in Machine Learning. Video of the talk. (Feb 2022)
- A Rigorous Study of Integrated Gradients Method and Extensions to Internal Neuron Attributions is available on arXiv. Joint work with Daniel Lundstrom and Tianjian Huang (Feb 2022).
- We launched the USC-Meta Center for Research and Education in AI and Learning (January 2022)
- Presented some of our group works in One World Optimization Seminar. YouTube video (January 2022)
- I am honored that my research proposal is selected for AFOSR Young Investigator Research Program (YIP) Award (November 2021).
- DAIR: Data augmented invariant regularization is available on arXiv. Joint work with Tianjian Huang, Chinnadhurai Sankar, Pooyan Amini, Satwik Kottur, Alborz Geramifard, and Ahmad Beirami (October 2021).
- Nonconvex-Nonconcave Min-Max Optimization with a Small Maximization Domain is available on arXiv. Joint work with Babak Barazandeh and Dmitrii Ostrovskii (October 2021).
- RIFLE is available on GitHub! RIFLE is a large-scale data imputation package designed based on a distributionally robust optimization framework. Check RIFLE package and the paper! (September 2021).
- Dmitrii Ostrovskii joined the Mathematics department at the University of Southern California as an Assistant Professor of Mathematics (August 2021).
- Tianjian Huang is doing an internship at Facebook AI (June-December 2021).
- Sina Baharlouei is doing an internship at the Bosch Center for Artificial Intelligence (June-October 2021).
- FERMI: Fair empirical risk minimization via exponential Rényi mutual information accepted to ICML-21 Workshop on Socially Responsible Machine Learning for oral presentation. Joint work with Andrew Lowy, Rakesh Pavan Sina Baharlouei, and Ahmad Beirami (July 2021).
- DAIR: Data augmented invariant regularization accepted to ICML-21 Workshop on Uncertainly and Robustness in Deep Learning. Joint work with Tianjian Huang, Chinnadhurai Sankar, Pooyan Amini, Satwik Kottur, Alborz Geramifard, and Ahmad Beirami (July 2021).
- Efficient Search of First-Order Nash Equilibria in Nonconvex-Concave Smooth Min-Max Problems accepted to SIAM Journal on Optimization. Joint work with Dmitrii M. Ostrovskii and Andrew Lowy (July 2021).
- Linearized ADMM Converges to Second-Order Stationary Points for Non-Convex Problems published in IEEE Transactions on Signal Processing. Joint work with Songtao Lu, Jason Lee, and Mingyi Hong (July 2021).
- Locally Differentially Private Federated Learning: Efficient Algorithms with Tight Risk Bounds is available on arXiv. Joint work with Andrew Lowy (June 2021).
- Efficient Algorithms for Estimating the Parameters of Mixed Linear Regression Models is available on arXiv. Joint work with Babak Barazandeh, Ali Ghafelebashi, and Ram Sriharsha (May 2021).
- Meisam Razaviyayn won the 3M’s Non-Tenured Faculty Award (3M NTFA), 2021 (May 2021).
- Ahmad Beirami, Emily Black, Krishna Gummadi, Hoda Heidari, Baharan Mirzasoleiman, Meisam Razaviyayn, and Joshua Williams co-organized the Responsible AI Workshop in ICLR (April 2021).
- Alternating direction method of multipliers for quantization” is published in AISTATS 2021. A joint work with Tianjian Huang, Prajwal Singhania, Maziar Sanjabi, and Pabitra Mitra (April 2021).
- Congestion Reduction via Personalized Incentives is available online. Joint work with Ali Ghafelebashi and Maged Dessouky (April 2021).