About Me

Hi! I’m Sebastian Curi, I’m a Ph.D. student at ETH Zurich, under the supervision of Andreas Krause. Currently, I’m doing an internship at Google, under the supervision of Gabriel Dulac-Arnold and Marcin Andrychowicz. Previously, I did a stay at Gergely’s Neu Lab in Universtat Pompeu Fabra.

I’m interested in the intersection between theory and practice of Reinforcement Learning/Control Systems. In particular, I focus on how to bring these algorithms into practical methods. I also care about the robustness, safety, sample-efficiency, and risk-aversity (to name some aspects) of data-driven decision-making.

My Publications

  1. Curi, S., Bogunovic, I., & Krause, A. (2021). Combining Pessimism with Optimism for Robust and Efficient Model-Based Deep Reinforcement Learning. International Conference on Machine Learning (ICML).
  2. Treven, L., Curi, S., Mutny, M., & Krause, A. (2021). Learning Controllers for Unstable Linear Quadratic Regulators from a Single Trajectory. Learning for Dynamics and Control (L4DC).
  3. Serrano, J. B., Curi, S., Krause, A., & Neu, G. (2021). Logistic Q-Learning. International Conference on Artificial Intelligence and Statistics, 3610–3618.
  4. Urpı́ Núria Armengol, Curi, S., & Krause, A. (2021). Risk-Averse Offline Reinforcement Learning. International Conference on Learning Representations. https://openreview.net/forum?id=TBIzh9b5eaz
  5. Curi, S., Berkenkamp, F., & Krause, A. (2020). Efficient Model-Based Reinforcement Learning through Optimistic Policy Search and Planning. Neural Information Processing Systems (NeurIPS), 33.
  6. Curi, S., Levy, K., Jegelka, S., Krause, A., & others. (2020). Adaptive sampling for stochastic risk-averse learning. Neural Information Processing Systems (NeurIPS), 33.
  7. Curi, S., Melchior, S., Berkenkamp, F., & Krause, A. (2020). Structured Variational Inference in Partially Observable Unstable Gaussian Process State Space Models. Learning for Dynamics and Control (L4DC), 147–157.
  8. Curi, S., Levy, K. Y., & Krause, A. (2019). Adaptive Input Estimation in Linear Dynamical Systems with Applications to Learning-from-Observations. 2019 IEEE 58th Conference on Decision and Control (CDC), 4115–4120.
  9. Borsos, Z., Curi, S., Levy, K. Y., & Krause, A. (2019). Online variance reduction with mixtures. International Conference on Machine Learning (ICML), 705–714.
  10. Fiducioso, M., Curi, S., Schumacher, B., Gwerder, M., & Krause, A. (2019). Safe contextual Bayesian optimization for sustainable room temperature PID control tuning. International Joint Conference on Artificial Intelligence, (IJCAI-19), 5850–5856.
  11. Curi, S., Groß, D., & Dörfler, F. (2017). Control of low-inertia power grids: A model reduction approach. 2017 IEEE 56th Annual Conference on Decision and Control (CDC), 5708–5713.
  12. Bright, L. Z., Handley, M., Chien, I., Curi, S., Brownworth, L. A., D’hers, S., Bernier, U. R., Gurman, P., & Elman, N. M. (2016). Analytical models integrated with satellite images for optimized pest management. Precision Agriculture, 17(5), 628–636.
  13. Curi, S., Mas, I., & Pena, R. S. (2014). Autonomous flight of a commercial quadrotor. IEEE Latin America Transactions, 12(5), 853–858.
  14. Chi, A., Curi, S., Clayton, K., Luciano, D., Klauber, K., Alexander-Katz, A., D’hers, S., & Elman, N. M. (2014). Rapid Reconstitution Packages (RRPs) implemented by integration of computational fluid dynamics (CFD) and 3D printed microfluidics. Drug Delivery and Translational Research, 4(4), 320–333.