Generative adversarial network training is a continual learning problem

KJ Liang; L Carin

Generative adversarial network training is a continual learning problem

KJ Liang, L Carin

Research output: Contribution to journal › Article › peer-review

Abstract

Generative Adversarial Networks (GANs) have proven to be a powerful framework for learning to draw samples from complex distributions. However, GANs are also notoriously difficult to train, with mode collapse and oscillations a common problem. We hypothesize …

Original language	English (US)
Journal	arXiv preprint arXiv:1811.11083
State	Published - 2018
Externally published	Yes

Bibliographical note

Cited By (since 2018): 21

M1 - Query date: 2021-03-11 11:12:31

M1 - 21 cites: https://scholar.google.com/scholar?cites=4525626415891074742&as_sdt=2005&sciodt=0,5&hl=en

Access to Document

https://arxiv.org/abs/1811.11083

Cite this

@article{002ce80638764c01b7319abf59245555,

title = "Generative adversarial network training is a continual learning problem",

abstract = "Generative Adversarial Networks (GANs) have proven to be a powerful framework for learning to draw samples from complex distributions. However, GANs are also notoriously difficult to train, with mode collapse and oscillations a common problem. We hypothesize …",

author = "KJ Liang and L Carin",

note = "Cited By (since 2018): 21 M1 - Query date: 2021-03-11 11:12:31 M1 - 21 cites: https://scholar.google.com/scholar?cites=4525626415891074742&as_sdt=2005&sciodt=0,5&hl=en",

year = "2018",

language = "English (US)",

journal = "arXiv preprint arXiv:1811.11083",

publisher = "arxiv.org",

}

TY - JOUR

T1 - Generative adversarial network training is a continual learning problem

AU - Liang, KJ

AU - Carin, L

N1 - Cited By (since 2018): 21 M1 - Query date: 2021-03-11 11:12:31 M1 - 21 cites: https://scholar.google.com/scholar?cites=4525626415891074742&as_sdt=2005&sciodt=0,5&hl=en

PY - 2018

Y1 - 2018

N2 - Generative Adversarial Networks (GANs) have proven to be a powerful framework for learning to draw samples from complex distributions. However, GANs are also notoriously difficult to train, with mode collapse and oscillations a common problem. We hypothesize …

AB - Generative Adversarial Networks (GANs) have proven to be a powerful framework for learning to draw samples from complex distributions. However, GANs are also notoriously difficult to train, with mode collapse and oscillations a common problem. We hypothesize …

M3 - Article

JO - arXiv preprint arXiv:1811.11083

JF - arXiv preprint arXiv:1811.11083

ER -

Generative adversarial network training is a continual learning problem

Abstract

Bibliographical note

Access to Document

Fingerprint

Cite this