Bridge robust control tools with machine learning research (stochastic optimization, deep RL, etc)
Years at WID2016 - present
Ph.D., Aerospace Engineering, University of Minnesota–Twin Cities
Research DescriptionBin Hu has broad interests in bridging the techniques used in the control and machine learning communities. He has been working on tailoring robust control theory (integral quadratic constraints, dissipation inequalities, jump system theory, etc) to study stochastic optimization methods (stochastic gradient, stochastic average gradient, SAGA, SVRG, SDCA, Katyusha momentum, etc) and their applications in related machine learning problems (logistic regression, SVM, deep learning, etc). He also is particularly interested in leveraging robust control tools with deep reinforcement learning.
- Bin Hu and Laurent Lessard, “Dissipativity theory for Nesterov's accelerated method,” Proceedings of the 34th International Conference on Machine Learning, PMLR 70:1549-1557, 2017.
- Bin Hu, Peter Seiler, and Anders Rantzer, “A unified analysis of stochastic optimization methods using jump system theory and quadratic constraints,” Proceedings of the 2017 Conference on Learning Theory, PMLR 65:1157-1189, 2017.
- Hu, Bin, Márcio J. Lacerda, and Peter Seiler. "Robustness analysis of uncertain discrete‐time systems with dissipation inequalities and integral quadratic constraints." International Journal of Robust and Nonlinear Control 27(11): 1940-1962, 2017.
- Bin Hu and Peter Seiler, “Exponential decay rate conditions for uncertain linear systems using integral quadratic constraints,” IEEE Transactions on Automatic Control, 61(11): 3561-3567, 2016.
- Bin Hu and Peter Seiler, “Pivotal decomposition for reliability analysis of fault tolerant control systems on unmanned aerial vehicles,” Reliability Engineering & System Safety, vol. 140, pp.130-141, 2015.