Control & Learning Group, Carnegie Mellon University
Ph.D. students and postdoctoral researchers
Open the Google FormProspective students and postdocs
Join the research group
I recruit students and postdoctoral researchers for both the Control & Learning Group at Carnegie Mellon University in Pittsburgh and the Research and Development Center for Large Language Models at the National Institute of Informatics in Tokyo.
Before you apply
Get to know the research
The group works on control and learning for autonomous systems, from safety certificates and risk quantification to language-guided control and physical AI on humanoid robots. Here is the group website. Before you apply, I strongly suggest prospective students to read a few of my recent papers first to gauge whether your interests fit the group. The papers below are good starting points; the full list is on the publications page.
Recent papers at AI conferences
- Training-Free Guided Diffusion for Planning: A Unified Framework via Doob's h-Transform with Safety GuaranteesK. Hoshino, Y. S. Aluru, X. Deng, and Y. Nakahira · International Conference on Machine Learning (ICML), 2026
- Polynomial Convergence of Riemannian Diffusion ModelsX. Xu, Z. Zhang, Y. Nakahira, G. Qu, and Y. Chi · International Conference on Learning Representations (ICLR), 2026
- Online Adaptive Probabilistic Safety Certificate with Language GuidanceZ. Wang, H. Hoshino, X. Deng, and Y. Nakahira · Learning for Dynamics and Control Conference (L4DC), 2026
- Stabilizing Linear Systems under Partial Observability: Sample Complexity and Fundamental LimitsZ. Zhang, Y. Nakahira, and G. Qu · Conference on Neural Information Processing Systems (NeurIPS), 2025
- Safety Certificate Against Latent Variables with Partially Unidentifiable DynamicsH. Jing and Y. Nakahira · International Conference on Machine Learning (ICML), 2025
- Sample Complexity of Stabilizing LTI Systems on a Single Trajectory under Stochastic NoiseZ. Zhang, Y. Nakahira, and G. Qu · Conference on Uncertainty in Artificial Intelligence (UAI), 2025
- Physics-Informed Representation and Learning: Control and Risk QuantificationZ. Wang, R. Keller, X. Deng, K. Hoshino, T. Tanaka, and Y. Nakahira · AAAI Conference on Artificial Intelligence, 2024
- An Analytic Solution to Covariance Propagation in Neural NetworksO. Wright, Y. Nakahira, and J. M. F. Moura · International Conference on Artificial Intelligence and Statistics (AISTATS), 2024
Recent journal papers in control and learning
- Myopically Verifiable Probabilistic Certificates for Safe Control and LearningZ. Wang*, H. Jing*, C. Kurniawan, A. Chern, and Y. Nakahira · IEEE Transactions on Automatic Control (accepted)
- Generalizable Physics-Informed Learning for Stochastic Safety-Critical SystemsZ. Wang, A. Chern, and Y. Nakahira · IEEE Transactions on Automatic Control (accepted)
- Probabilistic Safety Certificate for Multi-Agent Safe Control and LearningH. Jing and Y. Nakahira · IEEE Transactions on Control of Network Systems (accepted)
- Physics-Informed Deep B-Spline Networks for Dynamical SystemsZ. Wang, R. Romagnoli, J. Ratchford, and Y. Nakahira · Journal of Machine Learning Research (JMLR), 2025
- Scalable Long-Term Safety Certificate for Large-Scale SystemsK. Hoshino, Z. Wang, and Y. Nakahira · IEEE Control Systems Letters, 2023
- Diversity-Enabled Sweet Spots in Layered Architectures and Speed–Accuracy Trade-Offs in Sensorimotor ControlY. Nakahira, Q. Liu, T. Sejnowski, and J. C. Doyle · Proceedings of the National Academy of Sciences, 2021
Google Forms
How to apply
Prospective students and postdocs interested in control, learning, and physical AI research are welcome to fill in the Google Form that matches your situation. Due to the large volume of inquiries, I cannot respond to individual inquiries by email, but I do review the Google Form submissions on a regular basis.
Control & Learning Group, Carnegie Mellon University
Current CMU undergraduate and master's students
Open the Google FormResearch and Development Center for Large Language Models, National Institute of Informatics
Students and postdoctoral researchers
Open the Google Form