Dear all,
This Friday (the 21st) at 4PM CET, John Harlim from Penn State will present his work on machine learning of missing dynamical systems at CWI. Here, the model error problem arising from missing dynamics is formulated as a supervised learning task using the Mori-Zwanzig formalism and Taken's embedding theorem. More info can be found in the abstract below. Feel free to share the zoom link if you think the talk may be of interest to your colleagues or students.
Kind regards,
Wouter Edeling
Join Zoom Meeting https://cwi-nl.zoom.us/j/86326565885?pwd=THBiY3BRSUkvVzQ1UjM4Y1RTNGhOZz09
Meeting ID: 863 2656 5885 Passcode: 931873
13 May 2021 16h00: John Harlim (Penn state): Machine learning of missing dynamical systems
In the talk, I will discuss a general closure framework to compensate for the model error arising from missing dynamical systems. The proposed framework reformulates the model error problem into a supervised learning task to estimate a very high-dimensional closure model, deduced from the Mori-Zwanzig representation of a projected dynamical system with projection operator chosen based on Takens embedding theory. Besides theoretical convergence, this connection provides a systematic framework for closure modeling using available machine learning algorithms. I will demonstrate numerical results using a kernel-based linear estimator as well as neural network-based nonlinear estimators. If time permits, I will also discuss error bounds and mathematical conditions that allow for the estimated model to reproduce the underlying stationary statistics, such as one-point statistical moments and auto-correlation functions, in the context of learning Ito diffusions.