Physics-Enhanced Machine Learning (with sampling)

Add to your list(s) Download to your calendar using vCal

• Iryna Burak (TUM)
• Thursday 03 October 2024, 15:00-16:00
• Centre for Mathematical Sciences, MR14.

If you have a question about this talk, please contact Matthew Colbrook.

I will present the recent work done by Felix Deitrich’s group in Munich. The talk will specifically focus on the SWIM method [1], a sampling algorithm that allows fast and accurate construction of neural network weights. I will cover the basic SWIM method and its recent developments: SWIM -PDE to solve partial differential equations [2] and SWIM -RNN to learn dynamical systems with a combination of neural networks and the Koopman operator.

[1] Bolager, E.L., IB, Datar, C., Sun, Q. and Dietrich, F., 2024. Sampling weights of deep neural networks. Advances in Neural Information Processing Systems, 36. [2] Datar, C., Kapoor, T., Chandra, A., Sun, Q., IB, Bolager, E.L., Veselovska, A., Fornasier, M. and Dietrich, F., 2024. Solving partial differential equations with sampled neural networks. arXiv preprint arXiv:2405.20836.

This talk is part of the Applied and Computational Analysis series.

This talk is included in these lists:

• All CMS events
• All Talks (aka the CURE list)
• Applied and Computational Analysis
• CMS Events
• Centre for Mathematical Sciences, MR14
• DAMTP info aggregator
• Featured lists
• Interested Talks
• My seminars
• Type the title of a new list here
• bld31

Note that ex-directory lists are not shown.