BEGIN:VCALENDAR
VERSION:2.0
METHOD:PUBLISH
CALSCALE:GREGORIAN
PRODID:-//WordPress - MECv6.5.6//EN
X-ORIGINAL-URL:https://cmc.deusto.eus/
X-WR-CALNAME:cmc.deusto.eus
X-WR-CALDESC:DeustoCCM - Chair of Computational Mathematics at University of Deusto
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-PUBLISHED-TTL:PT1H
X-MS-OLK-FORCEINSPECTOROPEN:TRUE
BEGIN:VEVENT
CLASS:PUBLIC
UID:MEC-ce76bd604997d26b8af5561c84f25517@cmc.deusto.eus
DTSTART:20240529T073000Z
DTEND:20240530T160000Z
DTSTAMP:20251031T212900Z
CREATED:20251031
LAST-MODIFIED:20251031
PRIORITY:5
TRANSP:OPAQUE
SUMMARY:ML&OC24: Machine Learning and Optimal Control summer school
DESCRIPTION:On May 29-30, 2024, Prof. Enrique Zuazua ( http://dcn.nat.fau.eu/zuazua/ ) is giving a course on “Control and Machine Learning” at the ML&OC24 summer school “Machine Learning and Optimal Control” on May 27-31, 2024 at Gaeta (Italy).\nAbstract. In this course we shall present some recent results on the interplay between control and Machine Learning, and more precisely, Supervised Learning, Universal Approximation and Normaliying flows. We adopt the perspective of the simultaneous or ensemble control of systems of Residual Neural Networks (ResNets). Roughly, each item to be classified corresponds to a different initial datum for the Cauchy problem of the ResNets, leading to an ensemble of solutions to be driven to the corresponding targets, associated to the labels, by means of the same control. \nWe present a genuinely nonlinear and constructive method, allowing to show that such an ambitious goal can be achieved, estimating the complexity of the control strategies. This property is rarely fulfilled by the classical dynamical systems in Mechanics and the very nonlinear nature of the activation function governing the ResNet dynamics plays a determinant role. It allows deforming half of the phase space while the other half remains invariant, a property that classical models in mechanics do not fulfill. The turnpike property is also analyzed in this context, showing that a suitable choice of the cost functional used to train the ResNet leads to more stable and robust dynamics.\nThis lecture is inspired in joint work, among others, with Borjan Geshkovski (MIT), Domènec Ruiz-Balet (IC, London), Martin Hernandez (FAU)  and Antonio Alvarez (UAM)\nCourses\n6 hours/course\n• Prof. Enrique Zuazua, FAU. Friedrich-Alexander-Universität Erlangen-Nürnberg (Germany)\nControl and Machine Learning\n• Prof. Lars Grüne, Universität Bayreuth (Germany)\nHigh-dimensional (optimal) feedback control with neural networks\n• Prof. Huyen Pham, Université Paris Cité (France)\nDeep (reinforcement) learning methods for stochastic control PDEs \nSchedule of the Course\nOrganizers\n• Alessandro Alla. Università Ca` Foscari Venezia\n• Elisabetta Carlini. Sapienza, Università di Roma\n• Giovanni Colombo. Università di Padova\n• Michele Palladino. Università de L’Aquila\n• Cristina Pignotti. Università de L’Aquila\nWHEN\nClasses by Prof. Zuazua: Wed.-Thu. May 29 – 30, 2024\nWed. May 29, 2024\n11.30H – 13:00H. Class 1. Control and Machine Learning\n17.00H – 18:30H. Class 2. Control and Machine Learning\nThu. May 30, 2024\n09.30H – 11:00H. Class 1. Control and Machine Learning\n17.00H – 18:30H. Class 2. Control and Machine Learning\nRegistration\nTo register, fill out the registration form\nSee more information about Venue and registration\nWHERE\nHotel Serapo\nSpiaggia di Serapo, Via Firenze, 11, 04024\nGaeta (Italy)\nFor more information, please check the official site of the event.\n_\nDon’t miss out our Upcoming events!\n
URL:https://cmc.deusto.eus/events-calendar/mloc24-machine-learning-and-optimal-control-summer-school/
CATEGORIES:EZuazua,School,Seminar/Talk,Workshop
ATTACH;FMTTYPE=image/png:https://cmc.deusto.eus/wp-content/uploads/MLandOC2024_EZuazua_29-30may2024.png
END:VEVENT
END:VCALENDAR