![]() |
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 > Machine Learning Reading Group @ CUED > Flow matching, stochastic interpolants and everything in between
![]() Flow matching, stochastic interpolants and everything in betweenAdd to your list(s) Download to your calendar using vCal
If you have a question about this talk, please contact Isaac Reid. Zoom link available upon request (it is sent out on our mailing list, eng-mlg-rcc [at] lists.cam.ac.uk). Sign up to our mailing list for easier reminders via lists.cam.ac.uk. Flow matching is the latest development in deep generative modelling and has already been applied to numerous tasks including protein design, image inpainting etc. Flow matching brought continuous normalising flows to the front stage, showing that they can be trained ‘simulation-free’, as in without solving an ODE at each training step. These are closely related to diffusion models, yet yielding noiseless trajectories at sampling. We will introduce relevant core ideas, and discuss the advantage and inconvenience of noisy trajectories. Suggested reading: 1. Lipman, Chen & Ben-Hamu et al. (2022) Flow Matching for Generative Modeling. URL : http://arxiv.org/abs/2210.02747. 2. Tong, Malkin & Huguet et al. (2023) Improving and Generalizing Flow-Based Generative Models With Minibatch Optimal Transport. URL : http://arxiv.org/abs/2302.00482. 3. Albergo, Boffi & Vanden-Eijnden (2023) Stochastic Interpolants: a Unifying Framework for Flows and Diffusions. URL : http://arxiv.org/abs/2303.08797 This talk is part of the Machine Learning Reading Group @ CUED series. This talk is included in these lists:
Note that ex-directory lists are not shown. |
Other listsStephen Cowley's Meetings Europe Research Seminar Series Empowered EmployabilityOther talksThe dissipation of DNA on small length scales: measuring the bending friction constant of dsDNA and consequences for looping time predictions Roundtable on Hegel's World Revolutions From Polymers to Biology and Beyond Neuroimmune signalling at and across brain borders regulating neuronal fate |