<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class="">
<div dir="auto" style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class="">
Dear all, <br class="">
<br class="">
This is Xiaodan Zhou from Analysis on metric spaces unit at Okinawa Institute of Science and Technology Graduate University. <br class="">
<br class="">
The next talk of our seminar is scheduled on <b class="">this coming Friday, July 16th 9 am. </b><br class="">
<br class="">
<b class="">Speaker</b>: Elisa Negrini, Worcester Polytechnic Institute<br class="">
<br class="">
<b class="">Title</b>: System identification through Lipschitz regularized deep neural networks<br class="">
<br class="">
<b class="">Abstract</b>: In this work we use neural networks to learn governing equations from data. Specifically, we reconstruct the right-hand side of a system of ODEs ẋ=f(t,x(t)) directly from observed uniformly time-sampled data using a neural network.
In contrast with other neural network-based approaches to this problem, we add a Lipschitz regularization term to our loss function. In the synthetic examples we observed empirically that this regularization results in a smoother approximating function and
better generalization properties when compared with non-regularized models, both on trajectory and non-trajectory data, especially in presence of noise. In contrast with sparse regression approaches, since neural networks are universal approximators, we do
not need any prior knowledge on the ODE system. Since the model is applied component wise, it can handle systems of any dimension, making it usable for real-world data. I will talk about the advantages and limitations of our method and propose possible future
directions of research.<br class="">
<br class="">
(The beginning part of the talk will be an introduction of neural networks and some well-known theorems. The talk will be accessible to a general audience.)<br class="">
<br class="">
You can obtain a zoom link through the following registration page:
<div class=""><a href="https://oist.zoom.us/meeting/register/tJAkduirqjwpHNzpxRgm7t21u4n4fajQU0vT" class="">https://oist.zoom.us/meeting/register/tJAkduirqjwpHNzpxRgm7t21u4n4fajQU0vT</a></div>
<div class=""><br class="">
More information of the seminar can be found here:<br class="">
<a href="https://groups.oist.jp/aoms/2021-summer-analysis-metric-spaces-seminar-0" class="">https://groups.oist.jp/aoms/2021-summer-analysis-metric-spaces-seminar-0</a><br class="">
<br class="">
Please feel free to contact me if there are any questions. <br class="">
<br class="">
Best regards,<br class="">
Xiaodan Zhou<br class="">
-------------------------------<br class="">
<br class="">
Analysis on Metric Spaces Unit<br class="">
Okinawa Institute of Science and Technology Graduate University<br class="">
Okinawa 904-0495, Japan<br class="">
xiaodan.zhou@oist.jp<br class="">
<br class="">
https://groups.oist.jp/aoms</div>
</div>
</body>
</html>