About

Mario Sznaier is the Dennis Picard Trustee Professor of Electrical and Computer Engineering at Northeastern University. He received his Ingeniero Electronico and Ingeniero en Sistemas de Computacion degrees from the Universidad de la Republica in Uruguay, and his MSEE and Ph.D. degrees from the University of Washington. His academic career includes positions as an Assistant Professor at the University of Central Florida, and later at Pennsylvania State University, where he was promoted to Associate Professor and then to Professor of Electrical Engineering. In July 2006, he joined Northeastern University’s Electrical and Computer Engineering Department. His research interests encompass dynamics-enabled machine learning, robust control, control-oriented identification, semi-algebraic optimization, and dynamic computer vision. Sznaier has held visiting appointments at the California Institute of Technology and has served as an adjunct professor at Penn State. He is actively involved in the academic community as an Associate Editor for the journal Automatica, Editor in Chief for the section on AI and Machine Learning Control of Frontiers in Control Engineering, and Chair of IFAC's Technical Committee on Robust Control. Recognized as an IEEE Fellow, he has received the IEEE Control Systems Society Distinguished Member Award. His work includes leading multidisciplinary research projects funded by agencies such as the National Science Foundation, Department of Defense, and Air Force Office of Scientific Research, focusing on areas like verifiable robust AI, epidemic control, cyber-physical systems, and human behavioral modeling.

Research topics

• Artificial Intelligence
• Computer Science
• Computer vision
• Mathematics
• Algorithm
• Mathematical optimization

Selected publications

• On incremental and semi-global exponential stability of gradient flows satisfying generalized Łojasiewicz inequalities

  ArXiv.org · 2026-03-26

  articleOpen access

  The Łojasiewicz inequality characterizes objective-value convergence along gradient flows and, in special cases, yields exponential decay of the cost. However, such results do not directly give rates of convergence in the state. In this paper, we use contraction theory to derive state-space guarantees for gradient systems satisfying generalized Łojasiewicz inequalities. We first show that, when the objective has a unique strongly convex minimizer, the generalized Łojasiewicz inequality implies semi-global exponential stability; on arbitrary compact subsets, this yields exponential stability. We then give two curvature-based sufficient conditions, together with constraints on the Łojasiewicz rate, under which the nonconvex gradient flow is globally incrementally exponentially stable.

  PublisherOA PDF

• Physics-Informed System Identification Using Randomized Atomic Features

  2026-01-01

  articleOpen access
  PublisherDOI

• On incremental and semi-global exponential stability of gradient flows satisfying generalized Łojasiewicz inequalities

  arXiv (Cornell University) · 2026-03-26

  preprintOpen access

  The Łojasiewicz inequality characterizes objective-value convergence along gradient flows and, in special cases, yields exponential decay of the cost. However, such results do not directly give rates of convergence in the state. In this paper, we use contraction theory to derive state-space guarantees for gradient systems satisfying generalized Łojasiewicz inequalities. We first show that, when the objective has a unique strongly convex minimizer, the generalized Łojasiewicz inequality implies semi-global exponential stability; on arbitrary compact subsets, this yields exponential stability. We then give two curvature-based sufficient conditions, together with constraints on the Łojasiewicz rate, under which the nonconvex gradient flow is globally incrementally exponentially stable.

  PublisherDOI

• Unsafe probabilities and risk contours for stochastic processes using convex optimization

  Automatica · 2026-03-31

  articleOpen accessSenior author

  When evaluating safety specifications for trajectories of a dynamical system, it is vital to be able to bound the worst-case probability of unsafety (constraint violation) Certifications of stochastic safety and worst-case probabilities of unsafety can be expressed as infinite-dimensional linear programs (e.g. stochastic barrier functions, occupation measure problems) This paper proves that the infinite-dimensional linear programs and their finite-dimensional Moment-Sum-of-Squares truncations are nonconservative (to the true probability of unsafety) under compactness and regularity conditions in stochastic dynamics. Unsafe-probability estimates and risk contours are generated for example stochastic processes.

  PublisherDOI

• Real-Time Adaptive Motion Planning via Point Cloud-Guided, Energy-Based Diffusion and Potential Fields

  IEEE Robotics and Automation Letters · 2025-07-23

  articleOpen accessSenior author

  Motivated by the problem of pursuit-evasion, we present a motion planning framework that combines energy-based diffusion models with artificial potential fields for robust real time trajectory generation in complex environments. Our approach processes obstacle information directly from point clouds, enabling efficient planning without requiring complete geometric representations. The framework employs classifier-free guidance training and integrates local potential fields during sampling to enhance obstacle avoidance. In dynamic scenarios, the system generates initial trajectories using the diffusion model and continuously refines them through potential field-based adaptation, demonstrating effective performance in pursuit-evasion scenarios with partial pursuer observability.

  PublisherOA PDFDOI

• Robust Data-Driven Receding Horizon Control

  ArXiv.org · 2025-10-07

  preprintOpen access

  This paper presents a data-driven receding horizon control framework for discrete-time linear systems that guarantees robust performance in the presence of bounded disturbances. Unlike the majority of existing data-driven predictive control methods, which rely on Willem's fundamental lemma, the proposed method enforces set-membership constraints for data-driven control and utilizes execution data to iteratively refine a set of compatible systems online. Numerical results demonstrate that the proposed receding horizon framework achieves better contractivity for the unknown system compared with regular data-driven control approaches.

  PublisherOA PDFDOI

• Robust Data-Driven Receding-Horizon Control for LQR with Input Constraints

  ArXiv.org · 2025-10-07

  preprintOpen accessSenior author

  This letter presents a robust data-driven receding-horizon control framework for the discrete time linear quadratic regulator (LQR) with input constraints. Unlike existing data-driven approaches that design a controller from initial data and apply it unchanged throughout the trajectory, our method exploits all available execution data in a receding-horizon manner, thereby capturing additional information about the unknown system and enabling less conservative performance. Prior data-driven LQR and model predictive control methods largely rely on Willem's fundamental lemma, which requires noise-free data, or use regularization to address disturbances, offering only practical stability guarantees. In contrast, the proposed approach extends semidefinite program formulations for the data-driven LQR to incorporate input constraints and leverages duality to provide formal robust stability guarantees. Simulation results demonstrate the effectiveness of the method.

  PublisherOA PDFDOI

• Safe Control for Pursuit-Evasion With Density Functions

  IEEE Control Systems Letters · 2025-01-01

  articleSenior author

  This letter presents a density function based safe control synthesis framework for the pursuit-evasion problem. We extend safety analysis to dynamic unsafe sets by formulating a reach-avoid type pursuit-evasion differential game as a robust safe control problem. Using density functions and semi-algebraic sets, we derive sufficient conditions for weak eventuality and evasion, reformulating the problem into a convex sum-of-squares program solvable via standard semidefinite programming solvers. This approach avoids the computational complexity of solving the Hamilton-Jacobi-Isaacs partial differential equation, offering a scalable and efficient framework. Numerical simulations demonstrate the efficacy of the proposed method.

  PublisherDOI

• 3D-HGS: 3D Half-Gaussian Splatting^*

  2025-06-10 · 7 citations

  article

  Photo-realistic image rendering from 3D scene reconstruction has advanced significantly with neural rendering techniques. Among these, 3D Gaussian Splatting (3D-GS) outperforms Neural Radiance Fields (NeRFs) in quality and speed but struggles with shape and color discontinuities. We propose 3D Half-Gaussian (3D-HGS) kernels as a plug-and-play solution to address these limitations. Our experiments show that 3D-HGS enhances existing 3D-GS methods, achieving state-of-the-art rendering quality without compromising speed. More demos and code are available at https://lihaolin88.github.io/CVPR-2025-3DHGS.

  PublisherDOI

• Challenges in Model Agnostic Controller Learning for Unstable Systems

  IEEE Control Systems Letters · 2025-01-01 · 1 citations

  article1st authorCorresponding

  Model agnostic controller learning, for instance by direct policy optimization, has been the object of renewed attention lately, since it avoids a computationally expensive system identification step. Indeed, direct policy search has been empirically shown to lead to optimal controllers in a number of cases of practical importance. However, to date, these empirical results have not been backed up with a comprehensive theoretical analysis for general problems. In this paper we use a simple example to show that direct policy optimization is not directly generalizable to other seemingly simple problems. In such cases, direct optimization of a performance index can lead to unstable pole/zero cancellations, resulting in the loss of internal stability and unbounded outputs in response to arbitrarily small perturbations. We conclude the paper by analyzing several alternatives to avoid this phenomenon, suggesting some new directions in direct control policy optimization.

  PublisherDOI

Recent grants

• Robust Identification of a Class of Structured Systems with High Dimensional Outputs and Applications

  NSF · $415k · 2009–2014

• A Systems Theoretic Approach to Robust Active Vision

  NSF · $79k · 2006–2007

• Risk Adjusted Robust Control Theory and Applications

  NSF · $160k · 2005–2007

• CPS: Frontier: Collaborative Research: Data-Driven Cyberphysical Systems

  NSF · $270k · 2017–2022

• Robust Identification and Model Validation for a Class of Nonlinear Dynamic Systems and Applications

  NSF · $380k · 2014–2019

Frequent coauthors

• Robert R. Bitmead

  University of California, San Diego

  688 shared

• Marco Lovera

  Politecnico di Milano

  688 shared

• Warren E. Dixon

  University of Florida

  688 shared

• Kirsten Morris

  684 shared

• James Farrell

  University of Arizona

  679 shared

• John Lygeros

  ETH Zurich

  679 shared

• Dragan Nešić

  679 shared

• P Khargonekar

  Office of International Affairs

  679 shared

Awards & honors

• IEEE Fellow
• IEEE Control Systems Society Distinguished Member Award