COOKIES: By using this website you agree that we can place Google Analytics Cookies on your device for performance monitoring. |
University of Cambridge > Talks.cam > Isaac Newton Institute Seminar Series > Neural Controlled Differential Equations
Neural Controlled Differential EquationsAdd to your list(s) Download to your calendar using vCal
If you have a question about this talk, please contact INI IT. This talk has been canceled/deleted Differential equations and neural networks are two of the most widespread modelling paradigms. I will talk about how to combine the best of both worlds through neural differential equations. Beginning with a brief overview of the broader field of neural differential equations (neural ODEs being the most famous example), I will then discuss recent work on neural controlled differential equations (neural CDEs) specifically. Intuitively these are “continuous time RNNs”. These offer memory efficiency, the ability to handle irregular data, strong priors on model space, high capacity function approximation, and draw on a deep well of theory on both sides. They are state-of-the-art models for irregular time series, and if time allows I will briefly outline extensions for how they may be used to train neural SDEs. This talk is part of the Isaac Newton Institute Seminar Series series. This talk is included in these lists:This talk is not included in any other list Note that ex-directory lists are not shown. |
Other listsEPRG Energy and Environment (E&E) Series Michaelmas 2011 Arctic Talks Department of GeographyOther talksComputational foundations of morphogenomics Observations of stellar oscillations induced by hot Jupiters Stemcell International Seminar - Prof. Paul Riley. BHF Oxbridge Centre for Regenerative Medicine The real-time formation of the hippocampal cognitive map The Mechanics of Bioinspired Artificial Intelligence Larmor Lecture - Covid, Chaos and Climate: How mathematical models help to explain the universe |