About

Urbashi Mitra is the Gordon S. Marshall Chair in Engineering and a Professor of Electrical and Computer Engineering and Computer Science at the University of Southern California. She holds a Ph.D. in Electrical Engineering from Princeton University, earned in 1994, and both her B.S. and M.S. degrees from the University of California at Berkeley in 1987 and 1989, respectively. Her research has focused on problems in communication theory, information theory, and signal processing, with early work centered on wireless communication systems motivated by commercial applications. Her current research addresses fundamental questions at the boundaries of communication, estimation, and control for science, including joint design of communication, sensing, and control in networks. Key applications of her work include wireless body area networks for mobile health, spectrum sensing, resource allocation for cognitive radio, underwater autonomous vehicle networks, and microbial networks. She has also applied modern statistical methods such as sparse approximation and low-rank techniques to various communication and signal processing problems.

Research topics

• Computer Science
• Artificial Intelligence
• Mathematics
• Algorithm
• Telecommunications
• Mathematical optimization
• Real-time computing
• Microeconomics
• Biology
• Ecology
• Economics

Selected publications

• Cramér-Rao Bounds on Sparse-Diffuse Channel Estimation

  2026-04-21

  preprintSenior author

  Lower bounds on the estimation of a hybrid sparse-diffuse channel and parameters of the channel are provided. The Hybrid Atomic-Least-Squares (HALS) algorithm was designed to jointly estimate the sparse and diffuse components with a combined atomic and ℓ<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> regularization for this hybrid, mixed-channel model. The Cramér-Rao Bound analysis in this paper focuses on the estimation of the channel parameters, resulting in a bound on the aggregate channel. Numerical results via simulations on synthetic data validate the efficacy of the HALS estimation strategy and show the improved predictive ability of the CRB analysis for the performance of HALS versus previously considered bounds.

  PublisherDOI

• From Relative Entropy to Minimax: A Unified Framework for Coverage in MDPs

  ArXiv.org · 2026-01-17

  articleOpen access

  Targeted and deliberate exploration of state--action pairs is essential in reward-free Markov Decision Problems (MDPs). More precisely, different state-action pairs exhibit different degree of importance or difficulty which must be actively and explicitly built into a controlled exploration strategy. To this end, we propose a weighted and parameterized family of concave coverage objectives, denoted by $U_ρ$, defined directly over state--action occupancy measures. This family unifies several widely studied objectives within a single framework, including divergence-based marginal matching, weighted average coverage, and worst-case (minimax) coverage. While the concavity of $U_ρ$ captures the diminishing return associated with over-exploration, the simple closed form of the gradient of $U_ρ$ enables an explicit control to prioritize under-explored state--action pairs. Leveraging this structure, we develop a gradient-based algorithm that actively steers the induced occupancy toward a desired coverage pattern. Moreover, we show that as $ρ$ increases, the resulting exploration strategy increasingly emphasizes the least-explored state--action pairs, recovering worst-case coverage behavior in the limit.

  PublisherOA PDF

• Atomic Hybrid Sparse/Diffuse Channel Estimation and Cramér-Rao Bounds Analysis

  HAL (Le Centre pour la Communication Scientifique Directe) · 2026-04-17

  preprintSenior author

  In this paper, an atomic hybrid sparse/diffuse (aHSD) channel model in the frequency domain is proposed. Based on a structural analysis of the resolvable paths and diffuse scattering statistics, the Hybrid Atomic-Least-Squares (HALS) algorithm is designed to estimate sparse/diffuse components with a combined atomic and ℓ2 regularization. A theoretical analysis of the Lagrange dual problem is conducted, and the conditions required for primal and dual solutions are provided, supporting an off-the-grid delay-time estimator. The Cramér-Rao Bound (CRB) analysis in this paper focuses on the estimation of the channel parameters, resulting in a bound on the aggregate channel. Lower and upper bounds for the CRB on parameters are derived as functions of the minimum separations between frequency parameters. Numerical results via simulations on synthetic and real data validate the efficacy of the HALS estimation strategy and show the improved predictive ability of the CRB analysis for the performance of HALS versus previously considered bounds.

  Publisher

• Atomic Hybrid Sparse/Diffuse Channel Estimation and Cramér-Rao Bounds Analysis

  ArXiv.org · 2026-05-03

  articleOpen accessSenior author

  In this paper, an atomic hybrid sparse/diffuse (aHSD) channel model in the frequency domain is proposed. Based on a structural analysis of the resolvable paths and diffuse scattering statistics, the Hybrid Atomic-Least-Squares (HALS) algorithm is designed to estimate sparse/diffuse components with a combined atomic and $\ell_2$ regularization. A theoretical analysis of the Lagrange dual problem is conducted, and the conditions required for primal and dual solutions are provided, supporting an off-the-grid delay-time estimator. The Cramér--Rao Bound (CRB) analysis in this paper focuses on the estimation of the channel parameters, resulting in a bound on the aggregate channel. Lower and upper bounds for the CRB on parameters are derived as functions of the minimum separations between frequency parameters. Numerical results via simulations on synthetic and real data validate the efficacy of the HALS estimation strategy and show the improved predictive ability of the CRB analysis for the performance of HALS versus previously considered bounds.

  PublisherOA PDF

• $κ$-Explorer: A Unified Framework for Active Model Estimation in MDPs

  Open MIND · 2026-02-23

  preprint

  In tabular Markov decision processes (MDPs) with perfect state observability, each trajectory provides active samples from the transition distributions conditioned on state-action pairs. Consequently, accurate model estimation depends on how the exploration policy allocates visitation frequencies in accordance with the intrinsic complexity of each transition distribution. Building on recent work on coverage-based exploration, we introduce a parameterized family of decomposable and concave objective functions $U_κ$ that explicitly incorporate both intrinsic estimation complexity and extrinsic visitation frequency. Moreover, the curvature $κ$ provides a unified treatment of various global objectives, such as the average-case and worst-case estimation error objectives. Using the closed-form characterization of the gradient of $U_κ$, we propose $κ$-Explorer, an active exploration algorithm that performs Frank-Wolfe-style optimization over state-action occupancy measures. The diminishing-returns structure of $U_κ$ naturally prioritizes underexplored and high-variance transitions, while preserving smoothness properties that enable efficient optimization. We establish tight regret guarantees for $κ$-Explorer and further introduce a fully online and computationally efficient surrogate algorithm for practical use. Experiments on benchmark MDPs demonstrate that $κ$-Explorer provides superior performance compared to existing exploration strategies.

  DOI

• Sketched Column-based Matrix Approximation

  IEEE Transactions on Signal Processing · 2026-01-01

  articleSenior author

  A new, practical algorithm, fast Sketched Columnbased Matrix Approximation (fSCMA), is proposed for low–rank matrix approximation. fSCMA leverages randomly, but fully sampled columns combined with structural side information, to achieve efficient and accurate approximations. The algorithm leverages both matrix sketching and side information to reduce complexity. A theoretical spectral bound on the reconstruction error is derived, improving the error bound by a factor of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">n</i> (in terms of key parameters) compared to state-of-the-art algorithms (SoTA). Experimental results on synthetic data demonstrate that fSCMA achieves competitive performance relative to SoTA, validating theoretical bounds, while significantly reducing computational complexity. Additionally, fSCMA shows strong improvement over prior methods when applied to real data.

  PublisherDOI

• Generation–Establishment Tradeoffs Shape the Temporal Window of Recombinant Evolution

  bioRxiv (Cold Spring Harbor Laboratory) · 2026-04-09

  articleOpen access

  Recent viral coinfection experiments show that recombinant genomes are generated readily and depend strongly on infection timing and order, yet only a small fraction give rise to persistent lineages. We develop a hybrid deterministic–stochastic framework that resolves this discrepancy by coupling density-dependent recombinant generation with stochastic establishment of rare lineages. The resulting hazard of successful lineage formation is generically non-monotonic, increasing with parental abundance through enhanced generation while decreasing under competitive suppression, and exhibits a unique interior maximum in parental-density space. As parental populations evolve, their trajectory across this hazard landscape defines a sharply localized temporal window of evolutionary opportunity. These results reveal a general principle: evolutionary success is determined not only by intrinsic fitness, but by when variants arise within a dynamically changing ecological context.

  PublisherOA PDFDOI

• Atomic Hybrid Sparse/Diffuse Channel Estimation and Cramér-Rao Bounds Analysis

  arXiv (Cornell University) · 2026-05-03

  preprintOpen accessSenior author

  In this paper, an atomic hybrid sparse/diffuse (aHSD) channel model in the frequency domain is proposed. Based on a structural analysis of the resolvable paths and diffuse scattering statistics, the Hybrid Atomic-Least-Squares (HALS) algorithm is designed to estimate sparse/diffuse components with a combined atomic and $\ell_2$ regularization. A theoretical analysis of the Lagrange dual problem is conducted, and the conditions required for primal and dual solutions are provided, supporting an off-the-grid delay-time estimator. The Cramér--Rao Bound (CRB) analysis in this paper focuses on the estimation of the channel parameters, resulting in a bound on the aggregate channel. Lower and upper bounds for the CRB on parameters are derived as functions of the minimum separations between frequency parameters. Numerical results via simulations on synthetic and real data validate the efficacy of the HALS estimation strategy and show the improved predictive ability of the CRB analysis for the performance of HALS versus previously considered bounds.

  PublisherDOI

• Learning to Intervene: Optimized Soft Intervention Selection for Causal Discovery

  2026-04-21

  articleSenior author

  Understanding causal structure is fundamental for scientific discovery, yet most existing methods rely on observational data or passively use available interventional datasets. While these approaches have advanced the field, they do not actively target the most uncertain or poorly represented parts of the structure. We address this challenge by optimizing intervention design under soft interventions and integrating it into causal subgraph learning. A key ingredient is a utility function that balances reconstruction accuracy, exploration, and diversity: it highlights poorly estimated subgraphs, revisits under-explored distributions, and promotes diverse coverage to distinguish correlation from causation. Experiments show that optimized intervention selection enables more efficient and reliable structure learning than purely observational approaches and random-intervention baselines, underscoring the importance of principled intervention design in advancing causal discovery.

  PublisherDOI

• Block ModShift: Model Privacy via Dynamic Designed Shifts

  IEEE Journal on Selected Areas in Communications · 2026-01-01

  articleSenior author

  The problem of model privacy against an eavesdropper (Eve) in a distributed learning environment is investigated. The solution is found via evaluating the Fisher Information Matrix (FIM) for the model learning problem for Eve. Through a model shift design process, the eavesdropper’s FIM can be driven to singularity, yielding a provably hard estimation problem for Eve. Both a one-shot and multi-shot solution are designed. These two approaches require the sharing of a modest amount of information with the central server learning the global model. The multi-shot solution has time-varying shifts that prevent Eve from using the temporal correlation of the gradients to learn the shifts. We design a convergence test for Eve to determine if model updates have been tampered with. However, our shift strategies pass the test and thus the shifts are not detectable. The single-shot and multi-shot methods are compared against a noise injection scheme and shown to offer superior performance.

  PublisherDOI

Recent grants

• CIF: Small: Statistical Learning Methods for Communications, Sensing and Control in Actuated Wireless Networks

  NSF · $165k · 2020–2021

• Collaborative NETS-NECO: Wireless Underwater Multi-tiered Acoustic Networks (WUMAN)

  NSF · $250k · 2008–2012

• ITR: From Sensor Networks to Multimedia Systems: New Views on MIMO Channels

  NSF · $237k · 2003–2007

• CIF: Small: Learning, Optimization & Analysis for Biologically Inspired Community Networks

  NSF · $600k · 2023–2026

• CIF: Small: Modeling and Analysis of Microbial Signaling

  NSF · $506k · 2018–2023

Frequent coauthors

• Nicolò Michelusi

  Arizona State University

  48 shared

• Srinivas Yerramalli

  Qualcomm (United States)

  39 shared

• Marcos M. Vasconcelos

  Florida A&M University - Florida State University College of Engineering

  38 shared

• Marco Levorato

  35 shared

• Gaurav S. Sukhatme

  34 shared

• Dhruva Kartik

  University of Southern California

  27 shared

• Geoffrey A. Hollinger

  27 shared

• Wenyi Zhang

  University of Science and Technology of China

  26 shared

Education

• Ph.D., Electrical Engineering

  University of California, Los Angeles

  1992

• M.S., Electrical Engineering

  University of California, Los Angeles

  1988

• B.S., Electrical Engineering

  Indian Institute of Technology, Kanpur

  1985

Awards & honors

• 2024 IEEE Information Theory Society Aaron D. Wyner Distingu…
• 2024 Academia Europaea Foreign Member
• 2024 IEEE Signal Processing Society Distinguished Lecturer
• 2024 IEEE Signal Processing Society, Signal Processing for C…
• 2023 IEEE Communications Society Plenary Speaker, IEEE Inter…