Time-varying Normalizing Flow for Generative Modeling of Dynamical Signals

We develop a time-varying normalizing flow (TVNF) for explicit generative modeling of dynamical signals. Being explicit, it can generate samples of dynamical signals, and compute the likelihood of a (given) dynamical signal sample. In the proposed model, signal flow in the layers of the normalizing...

Full description

Saved in:
Bibliographic Details
Published in2022 30th European Signal Processing Conference (EUSIPCO) pp. 1492 - 1496
Main Authors Ghosh, Anubhab, Fontcuberta, Aleix Espuna, Abdalmoaty, Mohamed R.-H., Chatterjee, Saikat
Format Conference Proceeding
LanguageEnglish
Published EUSIPCO 29.08.2022
Subjects
Backpropagation
Computational modeling
Europe
Generative learning
neural networks
normalizing flows
Recurrent neural networks
Signal processing
Task analysis
Toy manufacturing industry
Online AccessGet full text

Cover

More Information
Summary:We develop a time-varying normalizing flow (TVNF) for explicit generative modeling of dynamical signals. Being explicit, it can generate samples of dynamical signals, and compute the likelihood of a (given) dynamical signal sample. In the proposed model, signal flow in the layers of the normalizing flow is a function of time, which is realized using an encoded representation that is the output of a recurrent neural network (RNN). Given a set of dynamical signals, the parameters of TVNF are learned according to maximum-likelihood approach in conjunction with gradient descent (backpropagation). Use of the proposed model is illustrated for a toy application scenario - maximum-likelihood based speech-phone classification task.
ISSN:2076-1465