Computing systems with modularized infrastructure for training generative adversarial networks

A technique for infrastructure, computer systems, applied in the field of machine learning, capable of solving problems that do not exist in a meaningful way

Pending Publication Date: 2019-05-14
GOOGLE LLC
View PDF1 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

As a third example challenge, techniques and procedures for assessing the quality of GANs do not currently exist in a meaningful way

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Computing systems with modularized infrastructure for training generative adversarial networks
  • Computing systems with modularized infrastructure for training generative adversarial networks
  • Computing systems with modularized infrastructure for training generative adversarial networks

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0022] overview

[0023] Example aspects of the present disclosure relate to computing systems that provide a modular infrastructure for training generative adversarial networks (GANs). For example, a modular infrastructure could include lightweight libraries designed to make training and evaluating GANs easy. Users can easily train GANs by interacting with and / or relying on the modular infrastructure.

[0024] According to one aspect of the present disclosure, the modular infrastructure may include multiple different code sets that handle operations at and within various stages of the GAN training process. The code set may be modular. That is, the code sets can be designed to exist independently, but can be combined easily and intuitively. Thus, users can adopt some or all code sets, or can replace a code set with a custom code set, while still generating viable combinations.

[0025] More specifically, in some implementations, the modular infrastructure may include a c...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The application relates to computing systems with modularized infrastructure for training generative adversarial networks. Example aspects of the present disclosure are directed to computing systems that provide a modularized infrastructure for training Generative Adversarial Networks (GANs). For example, the modularized infrastructure can include a lightweight library designed to make it easy totrain and evaluate GANs. A user can interact with and / or depend on the modularized infrastructure to easily train GANs. According to one aspect of the present disclosure, the modularized infrastructure can include a number of distinct sets of code that handle various stages of and operations within the GAN training process. The sets of code can be modular. That is, the sets of code can be designedto exist independently yet be easily and intuitively combinable. Thus, the user can employ some or all of the sets of code or can replace a certain set of code with a set of custom-code while still generating a workable combination.

Description

[0001] related application [0002] This application claims priority and benefit to U.S. Nonprovisional Patent Application No. 16 / 159,093, filed October 12, 2018, and U.S. Provisional Patent Application No. 62 / 582,142, filed November 6, 2017. US Nonprovisional Patent Application No. 16 / 159,093 and US Provisional Patent Application No. 62 / 582,142 are hereby incorporated by reference in their entirety. technical field [0003] This disclosure relates generally to machine learning. More specifically, the present disclosure relates to computing systems that provide a modularized infrastructure for training generative adversarial networks. Background technique [0004] Training a machine learning model such as a neural network usually involves defining a loss function that tells the model how close or far it is to its target. For example, image classification networks are often given a loss function that penalizes them for giving them the wrong classification; a network that mi...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06N20/00G06N3/08
CPCG06N3/082G06N3/084G06V10/82G06N3/047G06N3/045G06F18/2414G06F9/448G06N3/08G06F18/40
Inventor 乔尔·肖尔塞尔吉奥·瓜达拉马·科塔多
Owner GOOGLE LLC
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap