Time/place
Weekly or biweekly over the Autumn/Winter 2019.
Lecture plan
| Time/date/place | Speaker | Contents |
|---|---|---|
| Tues 24.09 at 10:15 in room 400 | Vegard Antun | General introduction to deep neural networks. |
| Tues 08.10 at 10:15 in room 1020 | Michael Floater | Yarotsky's paper (part 1) |
| Tues 15.10 at 10:15 in room 1020 | Michael Floater | Yarotsky's paper (part 2) |
| Tues 22.10 at 10:15 in room 400 | Jean Rabault |
Jean's papers |
| Tues 12.11 at 10:15 in room 621 | Alexander Lobbe | One or more of Weinan E's papers. |
Suggested papers
- Weinan E. A Proposal on Machine Learning via Dynamical Systems. Communications in Mathematics and Statistics, Volume 5, Issue 1, pp 1–11 (2017).
- Jiequn Han, Weinan E. Deep Learning Approximation for Stochastic Control Problems arXiv:1611.07422 (2016).
- Dmitry Yarotsky. Error bounds for approximations with deep ReLU networks. Neural Networks Volume 94, October 2017, Pages 103-114.
- Jim Magiera, Deep Ray, Jan S. Hesthaven and Christian Rohde. Constraint-Aware Neural Networks for Riemann Problems. arXiv:1904.12794, 2019.
- Kjetil O. Lye, Siddhartha Mishra and Deep Ray. Deep learning observables in computational fluid dynamics. arXiv:1903.03040
- Siddhartha Mishra. A machine learning framework for data driven acceleration of computations of differential equations. arXiv:1807.09519, 2018.
- Jean Rabault, Miroslav Kuchta, Atle Jensen, Ulysse Réglade and Nicolas Cerardi. Artificial neural networks trained through deep reinforcement learning discover control strategies for active flow control. Journal of Fluid Mechanics (865) pp 281-302, 2019.
J Rabault, A Kuhnle. Accelerating Deep Reinforcement Learning of Active Flow Control strategies through a multi-environment approach. Accepted, under production, Physics of Fluids (2019)
J Viquerat, J Rabault, A Kuhnle, H Ghraieb, E Hachem. Direct shape optimization through deep reinforcement learning. Submitted (2019) - Wang, Shen, Long and Dong. Learning to Discretize: Solving 1D Scalar Conservation Laws via Deep Reinforcement Learning. arXiv:1905.11079
- ...