UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Virtual Adversarial Ladder Networks For Semi-supervised Learning

Shinoda, S; Worrall, DE; Brostow, GJ; (2017) Virtual Adversarial Ladder Networks For Semi-supervised Learning. In: Proceedings of the NIPS 2017 LLD Workshop. NIPS 2017 LLD Workshop: Long Beach, CA, USA. Green open access

[img]
Preview
Text
Brostow_LLD_2017_paper_5.pdf - Published version

Download (2MB) | Preview

Abstract

Semi-supervised learning (SSL) partially circumvents the high cost of labeling data by augmenting a small labeled dataset with a large and relatively cheap unlabeled dataset drawn from the same distribution. This paper offers a novel interpretation of two deep learning-based SSL approaches, ladder networks and virtual adversarial training (VAT), as applying distributional smoothing to their respective latent spaces. We propose a class of models that fuse these approaches. We achieve near-supervised accuracy with high consistency on the MNIST dataset using just 5 labels per class: our best model, ladder with layer-wise virtual adversarial noise (LVAN-LW), achieves 1.42%±0.12 average error rate on the MNIST test set, in comparison with 1.62%±0.65 reported for the ladder network. On adversarial examples generated with L2-normalized fast gradient method, LVAN-LW trained with 5 examples per class achieves average error rate 2.4%±0.3 compared to 68.6%±6.5 for the ladder network and 9.9%±7.5 for VAT.

Type: Proceedings paper
Title: Virtual Adversarial Ladder Networks For Semi-supervised Learning
Event: NIPS 2017 LLD Workshop
Location: Long Beach, USA
Dates: 04 December 2017 - 09 December 2017
Open access status: An open access version is available from UCL Discovery
Publisher version: https://lld-workshop.github.io/2017/
Language: English
Additional information: This version is the version of record. For information on re-use, please refer to the publisher’s terms and conditions.
UCL classification: UCL
UCL > Provost and Vice Provost Offices
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science
URI: https://discovery.ucl.ac.uk/id/eprint/10039217
Downloads since deposit
4Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item