About

Shahin Shahrampour is currently an Assistant Professor in the Department of Mechanical & Industrial Engineering at Northeastern University. He previously served as an Assistant Professor in the Departments of Industrial & Systems Engineering and Electrical & Computer Engineering (by courtesy) at Texas A&M University from 2018 to 2021. Before his tenure at Texas A&M, he was a Postdoctoral Fellow in the School of Engineering and Applied Sciences at Harvard University. He holds a Ph.D. in Electrical and Systems Engineering from the University of Pennsylvania, along with a Master’s degree in Statistics from The Wharton School and a Master’s degree in Electrical Engineering from the University of Pennsylvania. His educational background also includes a Bachelor’s degree in Electrical Engineering from Sharif University of Technology. His research focuses on optimization and control, multi-agent systems, machine learning, and reinforcement learning. He has been recognized with several awards, including the NSF CAREER Award in 2025, the Martin W. Essigmann Outstanding Teaching Award in 2026, and the Best Paper Award at IEEE ICASSP in 2022. His work involves developing scalable, fast, and online decentralized manifold optimization in multi-agent networks, as well as advancing distributed optimization in non-convex environments with applications to networked machine learning. Shahrampour’s research aims to contribute to control and engineering applications through innovative approaches in decentralized and distributed optimization, with a focus on real-time learning and control processes.

Research topics

• Computer Science
• Artificial Intelligence
• Mathematics
• Pure mathematics
• Combinatorics
• Statistics
• Mathematical analysis
• Computer Security
• Geometry
• Discrete mathematics
• Engineering
• Applied mathematics
• Psychology
• Simulation
• Mathematical optimization

Selected publications

• Theoretical Analysis of Measure Consistency Regularization for Partially Observed Data

  Open MIND · 2026-02-01

  preprintSenior author

  The problem of corrupted data, missing features, or missing modalities continues to plague the modern machine learning landscape. To address this issue, a class of regularization methods that enforce consistency between imputed and fully observed data has emerged as a promising approach for improving model generalization, particularly in partially observed settings. We refer to this class of methods as Measure Consistency Regularization (MCR). Despite its empirical success in various applications, such as image inpainting, data imputation and semi-supervised learning, a fundamental understanding of the theoretical underpinnings of MCR remains limited. This paper bridges this gap by offering theoretical insights into why, when, and how MCR enhances imputation quality under partial observability, viewed through the lens of neural network distance. Our theoretical analysis identifies the term responsible for MCR's generalization advantage and extends to the imperfect training regime, demonstrating that this advantage is not always guaranteed. Guided by these insights, we propose a novel training protocol that monitors the duality gap to determine an early stopping point that preserves the generalization benefit. We then provide detailed empirical evidence to support our theoretical claims and to show the effectiveness and accuracy of our proposed stopping condition. We further provide a set of real-world data simulations to show the versatility of MCR under different model architectures designed for different data sources.

  DOI

• Theoretical Analysis of Measure Consistency Regularization for Partially Observed Data

  ArXiv.org · 2026-02-01

  articleOpen accessSenior author

  The problem of corrupted data, missing features, or missing modalities continues to plague the modern machine learning landscape. To address this issue, a class of regularization methods that enforce consistency between imputed and fully observed data has emerged as a promising approach for improving model generalization, particularly in partially observed settings. We refer to this class of methods as Measure Consistency Regularization (MCR). Despite its empirical success in various applications, such as image inpainting, data imputation and semi-supervised learning, a fundamental understanding of the theoretical underpinnings of MCR remains limited. This paper bridges this gap by offering theoretical insights into why, when, and how MCR enhances imputation quality under partial observability, viewed through the lens of neural network distance. Our theoretical analysis identifies the term responsible for MCR's generalization advantage and extends to the imperfect training regime, demonstrating that this advantage is not always guaranteed. Guided by these insights, we propose a novel training protocol that monitors the duality gap to determine an early stopping point that preserves the generalization benefit. We then provide detailed empirical evidence to support our theoretical claims and to show the effectiveness and accuracy of our proposed stopping condition. We further provide a set of real-world data simulations to show the versatility of MCR under different model architectures designed for different data sources.

  PublisherOA PDF

• Online Optimization on Hadamard Manifolds: Curvature-Independent Regret Bounds on Horospherically Convex Objectives

  IEEE Control Systems Letters · 2025-01-01

  articleSenior author
  PublisherDOI

• Online Optimization on Hadamard Manifolds: Curvature Independent Regret Bounds on Horospherically Convex Objectives

  ArXiv.org · 2025-09-14

  preprintOpen accessSenior author

  We study online Riemannian optimization on Hadamard manifolds under the framework of horospherical convexity (h-convexity). Prior work mostly relies on the geodesic convexity (g-convexity), leading to regret bounds scaling poorly with the manifold curvature. To address this limitation, we analyze Riemannian online gradient descent for h-convex and strongly h-convex functions and establish $O(\sqrt{T})$ and $O(\log(T))$ regret guarantees, respectively. These bounds are curvature-independent and match the results in the Euclidean setting. We validate our approach with experiments on the manifold of symmetric positive definite (SPD) matrices equipped with the affine-invariant metric. In particular, we investigate online Tyler's $M$-estimation and online Fréchet mean computation, showing the application of h-convexity in practice.

  PublisherOA PDFDOI

• ADARL: Adaptive Low-Rank Structures for Robust Policy Learning under Uncertainty

  ArXiv.org · 2025-10-13

  preprintOpen access

  Robust reinforcement learning (Robust RL) seeks to handle epistemic uncertainty in environment dynamics, but existing approaches often rely on nested min--max optimization, which is computationally expensive and yields overly conservative policies. We propose \textbf{Adaptive Rank Representation (AdaRL)}, a bi-level optimization framework that improves robustness by aligning policy complexity with the intrinsic dimension of the task. At the lower level, AdaRL performs policy optimization under fixed-rank constraints with dynamics sampled from a Wasserstein ball around a centroid model. At the upper level, it adaptively adjusts the rank to balance the bias--variance trade-off, projecting policy parameters onto a low-rank manifold. This design avoids solving adversarial worst-case dynamics while ensuring robustness without over-parameterization. Empirical results on MuJoCo continuous control benchmarks demonstrate that AdaRL not only consistently outperforms fixed-rank baselines (e.g., SAC) and state-of-the-art robust RL methods (e.g., RNAC, Parseval), but also converges toward the intrinsic rank of the underlying tasks. These results highlight that adaptive low-rank policy representations provide an efficient and principled alternative for robust RL under model uncertainty.

  PublisherOA PDFDOI

• Decentralized Online Riemannian Optimization Beyond Hadamard Manifolds

  ArXiv.org · 2025-09-09

  preprintOpen accessSenior author

  We study decentralized online Riemannian optimization over manifolds with possibly positive curvature, going beyond the Hadamard manifold setting. Decentralized optimization techniques rely on a consensus step that is well understood in Euclidean spaces because of their linearity. However, in positively curved Riemannian spaces, a main technical challenge is that geodesic distances may not induce a globally convex structure. In this work, we first analyze a curvature-aware Riemannian consensus step that enables a linear convergence beyond Hadamard manifolds. Building on this step, we establish a $O(\sqrt{T})$ regret bound for the decentralized online Riemannian gradient descent algorithm. Then, we investigate the two-point bandit feedback setup, where we employ computationally efficient gradient estimators using smoothing techniques, and we demonstrate the same $O(\sqrt{T})$ regret bound through the subconvexity analysis of smoothed objectives.

  PublisherOA PDFDOI

• Linear Convergence of Independent Natural Policy Gradient in Games With Entropy Regularization

  IEEE Control Systems Letters · 2024-01-01 · 2 citations

  articleSenior author

  This letter focuses on the entropy-regularized independent natural policy gradient (NPG) algorithm in multi-agent reinforcement learning. In this letter, agents are assumed to have access to an oracle with exact policy evaluation and seek to maximize their respective independent rewards. Each individual’s reward is assumed to depend on the actions of all agents in the multi-agent system, leading to a game between agents. All agents make decisions under a policy with bounded rationality, which is enforced by the introduction of entropy regularization. In practice, a smaller regularization implies that agents are more rational and behave closer to Nash policies. On the other hand, with larger regularization agents tend to act randomly, which ensures more exploration. We show that, under sufficient entropy regularization, the dynamics of this system converge at a linear rate to the quantal response equilibrium (QRE). Although regularization assumptions prevent the QRE from approximating a Nash equilibrium (NE), our findings apply to a wide range of games, including cooperative, potential, and two-player matrix games. We also provide extensive empirical results on multiple games (including Markov games) as a verification of our theoretical analysis.

  PublisherDOI

• Tracking Dynamic Gaussian Density with a Theoretically Optimal Sliding Window Approach

  Lecture notes in computer science · 2024-01-01

  book-chapterSenior author
  PublisherDOI

• Online Optimization Perspective on First-Order and Zero-Order Decentralized Nonsmooth Nonconvex Stochastic Optimization

  arXiv (Cornell University) · 2024-06-03

  preprintOpen accessSenior author

  We investigate the finite-time analysis of finding ($δ,ε$)-stationary points for nonsmooth nonconvex objectives in decentralized stochastic optimization. A set of agents aim at minimizing a global function using only their local information by interacting over a network. We present a novel algorithm, called Multi Epoch Decentralized Online Learning (ME-DOL), for which we establish the sample complexity in various settings. First, using a recently proposed online-to-nonconvex technique, we show that our algorithm recovers the optimal convergence rate of smooth nonconvex objectives. We then extend our analysis to the nonsmooth setting, building on properties of randomized smoothing and Goldstein-subdifferential sets. We establish the sample complexity of $O(δ^{-1}ε^{-3})$, which to the best of our knowledge is the first finite-time guarantee for decentralized nonsmooth nonconvex stochastic optimization in the first-order setting (without weak-convexity), matching its optimal centralized counterpart. We further prove the same rate for the zero-order oracle setting without using variance reduction.

  PublisherOA PDFDOI

• Linear Convergence of Independent Natural Policy Gradient in Games with Entropy Regularization

  arXiv (Cornell University) · 2024-05-04

  preprintOpen accessSenior author

  This work focuses on the entropy-regularized independent natural policy gradient (NPG) algorithm in multi-agent reinforcement learning. In this work, agents are assumed to have access to an oracle with exact policy evaluation and seek to maximize their respective independent rewards. Each individual's reward is assumed to depend on the actions of all the agents in the multi-agent system, leading to a game between agents. We assume all agents make decisions under a policy with bounded rationality, which is enforced by the introduction of entropy regularization. In practice, a smaller regularization implies the agents are more rational and behave closer to Nash policies. On the other hand, agents with larger regularization acts more randomly, which ensures more exploration. We show that, under sufficient entropy regularization, the dynamics of this system converge at a linear rate to the quantal response equilibrium (QRE). Although regularization assumptions prevent the QRE from approximating a Nash equilibrium, our findings apply to a wide range of games, including cooperative, potential, and two-player matrix games. We also provide extensive empirical results on multiple games (including Markov games) as a verification of our theoretical analysis.

  PublisherOA PDFDOI

Recent grants

• Collaborative Online Optimization for Efficient Model-Based Learning

  NSF · $500k · 2019–2021

Frequent coauthors

• Ali Jadbabaie

  36 shared

• T. T. Chang

  19 shared

• Youbang Sun

  Northeastern University

  17 shared

• Alexander Rakhlin

  16 shared

• Yinsong Wang

  16 shared

• Vahid Tarokh

  14 shared

• Shixiang Chen

  Chang'an University

  8 shared

• Alfredo García

  7 shared

Awards & honors

• Martin W. Essigmann Outstanding Teaching Award, 2026
• NSF CAREER Award, 2025
• Best Paper Award in IEEE Conference on Acoustics, Speech, an…
• TEES Engineering Genesis Award for Multidisciplinary Researc…