cv
Basics
| Name | Yahya Sattar |
| Label | Foundational AI Postdoctoral Fellow, Cornell University |
| ysattar@cornell.edu | |
| Phone | +1 (310) 801 8552 |
| Url | https://yahya-sattar.github.io |
| Summary | My research studies the fundamental limits of learning-based control, autoregressive prediction, and world modeling in nonlinear, uncertain, and partially observed settings. I develop theoretical foundations and efficient algorithms for learning nonlinear dynamics and latent representations from sequential data, and for using the learned dynamics and representations to predict, plan, and control. My work connects nonlinear system identification, adaptive control, reinforcement learning, autoregressive prediction and world modeling, drawing on tools from statistical learning, optimization, and control theory. |
Work
-
2026 - Present Foundational AI Postdoctoral Fellow
Cornell University, Department of Computer Science · Ithaca, NY, USA
Mentored by Prof. Sarah Dean and Prof. Maryam Fazel.
-
2023 - 2026 Postdoctoral Associate
Cornell University, Department of Computer Science · Ithaca, NY, USA
Hosted by Prof. Sarah Dean.
-
2015 - 2017 Research Assistant
LUMS University, Department of Electrical Engineering · Lahore, Pakistan
Hosted by Prof. Zubair Khalid.
Education
-
2017 - 2023 PhD in Electrical Engineering
University of California, Riverside · Riverside, CA, USA
Department of ECE · Advised by Prof. Samet Oymak
-
2017 - 2019 MS in Electrical Engineering
University of California, Riverside · Riverside, CA, USA
Department of ECE
-
2011 - 2015 BS in Electrical Engineering
LUMS University · Lahore, Pakistan
Department of EE, School of Science and Engineering
Awards
- 2026 – 2027
Foundational AI Postdoctoral Fellowship
Cornell University
- 2025
Best Paper Award Nominee
Conference on Learning for Dynamics & Control (L4DC)
- 2017 – 2018
Dean’s Distinguished Fellowship Award
UC Riverside
- 2011 – 2015
National Outreach Program (NOP) Scholarship Award
LUMS University
Publications
-
2027 Dual Control of Linear Systems from Bilinear Observations with Belief Space Model Predictive Control
In submission to the 2027 American Control Conference (ACC 2027, Senior Author)
D. Cao, B. Du, A. Lowitt, S. Choi, S. Dean, Y. Sattar
-
2027 Bilinear System Identification: A Unified Finite-Sample Analysis
In submission to the IFAC Journal of Systems and Control (Journal)
Y. Sattar, Y. Jedra, M. Fazel, S. Dean
-
2027 Performance Gaps Between Open-Loop, Certainty-Equivalent, and Optimal Control Under Action-Dependent Observations
In submission to the 30th International Conference on Artificial Intelligence and Statistics (AISTATS 2027)
L. M-Zhang, S. Choi, Y. Sattar, Y. Jedra, S. Dean, M. Fazel
-
2027 Extracting Algorithms in Pre-trained LLMs: A Case on Hidden Markov Models
In submission to the 15th International Conference on Learning Representations (ICLR 2027)
Y. Dai, Z. Gao, Y. Sattar, J.J. Sun, S. Dean
-
2026 Identification and Adaptive Control of Markov Jump Systems: Sample Complexity and Regret Bounds
In revision to the IEEE Transactions on Automatic Control (TAC, Journal)
Y. Sattar†, Z. Du†, D.A. Tarzanagh, L. Balzano, N. Ozay, S. Oymak
-
2026 Learning & Control Beyond Linearity: Towards a Non-asymptotic Theory for Bilinear Systems
The IEEE 65th Conference on Decision and Control (CDC 2026, Tutorial)
Y. Sattar, Y. Jedra, R. Strässer, F. Allgöwer, M. Fazel, S. Dean
-
2026 Explore-then-Commit for Nonstationary Linear Bandits with Latent Dynamics
The 29th International Conference on Artificial Intelligence and Statistics (AISTATS 2026)
S. Choi, Y. Sattar, Y. Jedra, M. Fazel, S. Dean
-
2026 Two-Layer Linear Auto-Regressive Models Estimate Latent States
The 43rd International Conference on Machine Learning (ICML 2026)
Y. Sattar, S. Choi, L. M-Zhang, Y. Jedra, M. Fazel, S. Dean
-
2025 A Case Study of Low Ranked Self-Expressive Structures in Neural Network Representations
The 2nd Conference on Parsimony and Learning, Stanford University (CPAL 2025, Oral)
U.S. Saini, W. Shiao, Y. Sattar, Y. Dahiya, S. Oymak, E.E. Papalexakis
-
2025 Learning Linear Dynamics from Bilinear Observations
The 2025 American Control Conference (ACC 2025)
Y. Sattar, Y. Jedra, S. Dean
-
2025 Sub-optimality of the Separation Principle for Quadratic Control from Bilinear Observations
IEEE 64th Conference on Decision and Control (CDC 2025)
Y. Sattar, S. Choi, Y. Jedra, M. Fazel, S. Dean
-
2025 Finite Sample Identification of Partially Observed Bilinear Dynamical Systems
The 7th Conference on Learning for Dynamics and Control, U Michigan Ann Arbor (L4DC 2025, Oral)
Y. Sattar†, Y. Jedra†, M. Fazel, S. Dean
-
2025 Pre-trained Large Language Models Learn to Predict Hidden Markov Models In-context
The 39th Annual Conference on Neural Information Processing Systems (NeurIPS 2025)
Y. Dai, Z. Gao, Y. Sattar, S. Dean, J.J. Sun
-
2024 Random Features Approximation for Control-Affine Systems
The 6th Conference on Learning for Dynamics and Control, U Oxford (L4DC 2024)
K. Kazemian, Y. Sattar, S. Dean
-
2022 Non-asymptotic and Accurate Learning of Nonlinear Dynamical Systems
Journal of Machine Learning Research (JMLR, Journal; poster at NeurIPS 2022)
Y. Sattar, S. Oymak
-
2022 Data-driven Control of Markov Jump Systems: Sample Complexity and Regret Bounds
The 2022 American Control Conference (ACC 2022)
Z. Du†, Y. Sattar†, D.A. Tarzanagh, L. Balzano, N. Ozay, S. Oymak
-
2022 Certainty Equivalent Quadratic Control for Markov Jump Systems
The 2022 American Control Conference (ACC 2022)
Y. Sattar†, Z. Du†, D.A. Tarzanagh, S. Oymak, L. Balzano, N. Ozay
-
2022 Finite Sample Identification of Bilinear Dynamical Systems
IEEE 61st Conference on Decision and Control (CDC 2022)
Y. Sattar, S. Oymak, N. Ozay
-
2021 Group Activity Recognition in Visual Data: A Retrospective Analysis of Recent Advancements
IEEE International Conference on Digital Futures and Transformative Technologies (ICoDT2 2021)
S. Sattar, Y. Sattar, M. Shahzad, M.M. Fraz
-
2021 Estimation of Groundwater Storage Variations in Indus River Basin Using GRACE Data
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2021)
Y. Sattar, Z. Khalid
-
2020 Convergence of Gradient-based Semi-supervised Learning with Self-training
arXiv (Preprint)
Y. Sattar, S. Oymak
-
2020 Exploring Weight Importance and Hessian Bias in Model Pruning
arXiv (Preprint)
M. Li, Y. Sattar, C. Thrampoulidis, S. Oymak
-
2020 Quickly Finding the Best Linear Model in High Dimensions via Projected Gradient Descent
IEEE Transactions on Signal Processing (TSP, Journal)
Y. Sattar, S. Oymak
-
2019 Accurate Reconstruction of Finite Rate of Innovation Signals on the Sphere
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2019)
Y. Sattar, Z. Khalid, R.A. Kennedy
-
2019 A Simple Framework for Learning Stabilizable Systems
IEEE Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP 2019)
Y. Sattar, S. Oymak
-
2017 Robust Reconstruction of Spherical Signals with Finite Rate of Innovation
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2017)
Y. Sattar, Z. Khalid, R.A. Kennedy
Talks
- CDC 2026
Learning and Control Beyond Linearity: Towards a Non-asymptotic Theory for Bilinear Systems
- Co-Organizer of the tutorial session at IEEE Conference on Decision and Control (2026).
- INFORMS 2026
Latent State Representation Learning in Auto-Regressive Models
- Invited talk at the INFORMS session on Recent Developments in Statistics and Learning, hosted by the Applied Probability Society at the 2026 INFORMS Annual Meeting (2026).
- 2026
Two-Layer Linear Auto-Regressive Models Estimate Latent States
- Selected talk at the Northeast Systems and Control Workshop, Princeton (2026).
- Poster at the Workshop on New Directions in Reinforcement Learning and Control, hosted by the NSF Institute for Mathematical and Statistical Innovation, U Chicago (2026).
- 2025
Finite Sample Identification of Partially Observed Bilinear Dynamical Systems
- Poster at the Workshop on Theoretical Foundations of Applied AI, hosted by the NSF Institute for Foundations of Data Science, UW Seattle (2025).
- Invited talk at Prof. Necmiye Ozay’s group meeting, U Michigan Ann Arbor (2025).
- 2024 – 2025
Learning & Control of Linear Dynamics from Bilinear Observations
- Invited talk at the Workshop on the Theoretical Aspects of Trustworthy AI, Simons Institute for Theory of Computing, UC Berkeley, presented by Prof. Sarah Dean (2025).
- Poster at the Northeast Systems and Control Workshop, U Pennsylvania (2024).
- Poster at the Princeton Workshop on Optimization, Learning, and Control (2024).
- Invited talk at the ICML Workshop on Foundations of Reinforcement Learning and Control, presented by Prof. Sarah Dean (2024).
- 2023
Statistical Learning Theory for Nonlinear Dynamical Systems
- Invited talk at Prof. Na Li’s group meeting, Harvard University (2023).
- Invited talk at Prof. Necmiye Ozay’s group meeting, U Michigan Ann Arbor (2023).
- 2022
Finite Sample Identification of Bilinear Dynamical Systems
- Selected talk at the 40th Southern California Control Workshop, Caltech (2022).
- 2021 – 2022
Identification and Adaptive Control of Markov Jump Systems: Sample Complexity and Regret Bounds
- Invited talk at the IEEE CDC Workshop on Statistical Learning Theory for Control (2022).
- Poster at the ICML Workshop on Reinforcement Learning Theory (2021).
- Selected talk at the 38th Southern California Control Workshop, UC Irvine (2021).
Teaching
- Winter 2023
Optimization for Machine Learning (EE/CS 248)
Teaching Assistant · University of California, Riverside, CA, USA
Graduate level course led by Prof. Samet Oymak.
- Spring 2020
Introduction to Deep Learning (EE/CS 228)
Teaching Assistant · University of California, Riverside, CA, USA
Graduate level course led by Prof. Samet Oymak.
- Spring 2019,
Winter 2020Probability, Random Variables, and Random Processes (EE 114)
Teaching Assistant · University of California, Riverside, CA, USA
Undergraduate level course led by Prof. Samet Oymak.
- Fall 2018,
Fall 2019Stochastic Processes (EE 215)
Teaching Assistant · University of California, Riverside, CA, USA
Graduate level course led by Prof. Samet Oymak.
Theses
- 2023
The Sample Complexity of Learning Dynamical Systems
PhD in Electrical Engineering, University of California, Riverside
- 2015
Self-Organizing Acoustic Localization Network
BS in Electrical Engineering, LUMS University
Mentoring
- 2024 – Present
Research Supervision
Cornell University, Ithaca NY, USA
Mentored multiple students in the CS department at Cornell. Research project topic: “Quadratic Control from Bilinear Observations”.
- Summer 2025
High School Outreach Program
Cornell University, Ithaca NY, USA
Research support volunteer for the 2025 high school outreach program at Cornell. Research project topic: “Control and Navigation of Weather Balloons”.
- Summer 2023
Guest Lecture
Cornell University, Ithaca NY, USA
Guest lecture for the course Machine Learning in Feedback Systems (CS 6784).
- 2024 – Present
Sunmook Choi
PhD student in Applied Mathematics, Cornell University
- 2025 – Present
Daniel Cao
BA+MEng student in Computer Science, Mathematics, and Statistics, Cornell University
- 2025 – Present
Beixi Du
MEng student in Financial Engineering, Cornell University
- 2026 – Present
Amanuel Adane
BS student in Computer Science, Cornell University
- 2025 – 2026
Andrew Lowitt
BS+MEng student in Computer Science and Mathematics, Cornell University
- 2025 – 2026
Beatriz de Lucio
MEng student in Computer Science, Cornell University
Service
- 2027
International Conference on Learning Representations (ICLR)
Conference Area Chair
- 2026
Conference on Neural Information Processing Systems (NeurIPS)
Conference Area Chair
- 2026
International Conference on Machine Learning (ICML)
Conference Area Chair
- 2022
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
Reviewer (Journal)
- 2021
IEEE Signal Processing Letters (SPL)
Reviewer (Journal)
- 2021,
2026International Conference on Artificial Intelligence and Statistics (AISTATS)
Reviewer (Conference)
- 2024,
2025Conference on Learning for Dynamics & Control (L4DC)
Reviewer (Conference)
- 2022
IEEE American Control Conference (ACC)
Reviewer (Conference)
- 2020
International Symposium on Information Theory (ISIT)
Reviewer (Conference)
Languages
| English | |
| Proficient |
| Urdu | |
| Native Speaker |
| Balti | |
| Native Speaker |
| Hindi | |
| Intermediate |
| Arabic | |
| Basic |
Interests
| Research Areas | ||||||
| Learning-based Control | ||||||
| Nonlinear System Identification | ||||||
| Adaptive Control & Reinforcement Learning | ||||||
| Autoregressive Prediction & World Modeling | ||||||
| Statistical Learning Theory & Optimization | ||||||
References
| Sarah Dean | |
| Assistant Professor, Computer Science, Cornell University Email: sdean@cornell.edu |
| Samet Oymak | |
| Associate Professor, EECS, University of Michigan Ann Arbor Email: oymak@umich.edu |
| Necmiye Ozay | |
| Professor, EECS, Robotics, University of Michigan Ann Arbor Email: necmiye@umich.edu |
| Laura Balzano | |
| Professor, EECS, Statistics, University of Michigan Ann Arbor Email: girasole@umich.edu |
| Maryam Fazel | |
| Professor, ECE, Mathematics, Statistics, Computer Science, University of Washington Seattle Email: mfazel@uw.edu |