WGAN and Optimal Transport

Add to your list(s) Download to your calendar using vCal

• Mark Rowland; Wenbo Gong
• Thursday 16 November 2017, 13:30-15:00
• Engineering Department, CBL Seminar Room 4-38.

If you have a question about this talk, please contact Alessandro Davide Ialongo.

Abstract

Optimal transport metrics are methods of comparing probability distributions, and are increasingly used in machine learning. We’ll give a general overview of these metrics, and discuss some of the statistical and computational issues with them when applying them in machine learning problems. The second part of the reading group will focus in on the Kantorovich-Rubenstein duality of Wasserstein distance and the Wasserstein GAN , a generative model introduced earlier this year that claims to fix some of the training difficulties associated with GANs, using an optimal transport distance in its training objective.

Recommended Reading

There is no required reading, although the WGAN paper (http://proceedings.mlr.press/v70/arjovsky17a/arjovsky17a.pdf) contains useful background material.

This talk is part of the Machine Learning Reading Group @ CUED series.

This talk is included in these lists:

• All Talks (aka the CURE list)
• Cambridge Centre for Data-Driven Discovery (C2D3)
• Cambridge Forum of Science and Humanities
• Cambridge Language Sciences
• Cambridge University Engineering Department Talks
• Cambridge talks
• Centre for Smart Infrastructure & Construction
• Chris Davis' list
• Computational Continuum Mechanics Group Seminars
• Engineering Department, CBL Seminar Room 4-38
• Featured lists
• Guy Emerson's list
• Hanchen DaDaDash
• Inference Group Journal Clubs
• Inference Group Summary
• Information Engineering Division seminar list
• Interested Talks
• ML
• Machine Learning Reading Group
• Machine Learning Reading Group @ CUED
• Machine Learning Summary
• Quantum Matter Journal Club
• Required lists for MLG
• School of Technology
• Simon Baker's List
• TQS Journal Clubs
• Trust & Technology Initiative - interesting events
• bld31
• custom
• ndk22's list
• ob366-ai4er
• rp587
• yk373's list

Note that ex-directory lists are not shown.