\dm_csml_event_details UCL ELLIS

f-GAN: Training Generative Neural Samplers using Variational Divergence Minimization


Speaker

Ryota Tomioka

Affiliation

Microsoft Research Cambridge

Date

Friday, 18 November 2016

Time

13:00-14:00

Location

Zoom

Link

Roberts Building 508

Event series

DeepMind/ELLIS CSML Seminar Series

Abstract

Generative neural samplers are probabilistic models that implement sampling using feedforward neural networks: they take a random input vector and produce a sample from a probability distribution defined by the network weights. These models are expressive and allow efficient computation of samples and derivatives, but cannot be used for computing likelihoods or for marginalization. The generative-adversarial training method allows to train such models through the use of an auxiliary discriminative neural network. We show that the generative-adversarial approach is a special case of an existing more general variational divergence estimation approach. We show that any f-divergence can be used for training generative neural samplers. We discuss the benefits of various choices of divergence functions on training complexity and the quality of the obtained generative models.

Biography