Generative Adversarial Networks

Gilad Cohen; Raja Giryes

doi:10.1007/978-3-031-24628-9_17

Generative Adversarial Networks

Gilad Cohen, Raja Giryes^*

^*Corresponding author for this work

School of Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Chapter › peer-review

3 Scopus citations

Abstract

Generative adversarial networks (GANs) are very popular frameworks for generating high-quality data and are immensely used in both the academia and industry in many domains. Arguably, their most substantial impact has been in the area of computer vision, where they achieve state-of-the-art image generation. This chapter gives an introduction to GANs, by discussing their principle mechanism and presenting some of their inherent problems during training and evaluation. We focus on these three issues: (1) mode collapse, (2) vanishing gradients, and (3) generation of low-quality images. We then list some architecture-variant and loss-variant GANs that remedy the above challenges. Lastly, we present two utilization examples of GANs for real-world applications: data augmentation and face images generation.

Original language	English
Title of host publication	Machine Learning for Data Science Handbook
Subtitle of host publication	Data Mining and Knowledge Discovery Handbook, Third Edition
Publisher	Springer International Publishing
Pages	375-400
Number of pages	26
ISBN (Electronic)	9783031246289
ISBN (Print)	9783031246272
DOIs	https://doi.org/10.1007/978-3-031-24628-9_17
State	Published - 1 Jan 2023

Access to Document

10.1007/978-3-031-24628-9_17

Cite this

@inbook{b16d8783f96a4796b746a297119dea1a,

title = "Generative Adversarial Networks",

abstract = "Generative adversarial networks (GANs) are very popular frameworks for generating high-quality data and are immensely used in both the academia and industry in many domains. Arguably, their most substantial impact has been in the area of computer vision, where they achieve state-of-the-art image generation. This chapter gives an introduction to GANs, by discussing their principle mechanism and presenting some of their inherent problems during training and evaluation. We focus on these three issues: (1) mode collapse, (2) vanishing gradients, and (3) generation of low-quality images. We then list some architecture-variant and loss-variant GANs that remedy the above challenges. Lastly, we present two utilization examples of GANs for real-world applications: data augmentation and face images generation.",

author = "Gilad Cohen and Raja Giryes",

note = "Publisher Copyright: {\textcopyright} Springer Nature Switzerland AG 2023.",

year = "2023",

month = jan,

day = "1",

doi = "10.1007/978-3-031-24628-9_17",

language = "אנגלית",

isbn = "9783031246272",

pages = "375--400",

booktitle = "Machine Learning for Data Science Handbook",

publisher = "Springer International Publishing",

}

TY - CHAP

T1 - Generative Adversarial Networks

AU - Cohen, Gilad

AU - Giryes, Raja

PY - 2023/1/1

Y1 - 2023/1/1

N2 - Generative adversarial networks (GANs) are very popular frameworks for generating high-quality data and are immensely used in both the academia and industry in many domains. Arguably, their most substantial impact has been in the area of computer vision, where they achieve state-of-the-art image generation. This chapter gives an introduction to GANs, by discussing their principle mechanism and presenting some of their inherent problems during training and evaluation. We focus on these three issues: (1) mode collapse, (2) vanishing gradients, and (3) generation of low-quality images. We then list some architecture-variant and loss-variant GANs that remedy the above challenges. Lastly, we present two utilization examples of GANs for real-world applications: data augmentation and face images generation.

AB - Generative adversarial networks (GANs) are very popular frameworks for generating high-quality data and are immensely used in both the academia and industry in many domains. Arguably, their most substantial impact has been in the area of computer vision, where they achieve state-of-the-art image generation. This chapter gives an introduction to GANs, by discussing their principle mechanism and presenting some of their inherent problems during training and evaluation. We focus on these three issues: (1) mode collapse, (2) vanishing gradients, and (3) generation of low-quality images. We then list some architecture-variant and loss-variant GANs that remedy the above challenges. Lastly, we present two utilization examples of GANs for real-world applications: data augmentation and face images generation.

UR - http://www.scopus.com/inward/record.url?scp=85183998224&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-24628-9_17

DO - 10.1007/978-3-031-24628-9_17

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.chapter???

AN - SCOPUS:85183998224

SN - 9783031246272

SP - 375

EP - 400

BT - Machine Learning for Data Science Handbook

PB - Springer International Publishing

ER -

Generative Adversarial Networks

Abstract

Access to Document

Other files and links

Fingerprint

Cite this