Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
畳み込みニューラルネットワークを使用したシーケンスの処理
Document Type and Number:
Japanese Patent JP6750121
Kind Code:
B2
Abstract:
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating of generating a neural network output from an input sequence. One of the methods includes, for each of the inputs, providing a current input sequence that comprises the input and the inputs preceding the input in the input sequence to a convolutional subnetwork comprising a plurality of dilated convolutional neural network layers, wherein the convolutional subnetwork is configured to, for each of the plurality of inputs: receive the current input sequence for the input, and process the current input sequence to generate an alternative representation for the input; and providing the alternative representations to an output subnetwork, wherein the output subnetwork is configured to receive the alternative representations and to process the alternative representations to generate the neural network output.

Inventors:
Aaron Herald Antonius van den Aalto
Thunder Etienne Rare Dealman
Nar Emmerich Karlbrenner
Karen Simonian
Oriol Vinals
Lasse Espeholt
Application Number:
JP2019533306A
Publication Date:
September 02, 2020
Filing Date:
September 06, 2017
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
Deep Mind Technologies Limited
International Classes:
G06N3/04; G06F40/40; G06F40/44; G10L15/16
Foreign References:
CA2810457A1
US20160063359
Other References:
Aaron van den Oord et al.,"Pixel Recurrent Neural Networks",arXiv.org [online],Cornell University,2016年 8月,arXiv:1601.06759v3,pp.1-11,[令和2年7月1日検索],インターネット
Aaron van den Oord et al.,"Conditional Image Generation with PixelCNN Decoders",arXiv.org [online],Cornell University,2016年 6月,arXiv:1606.05328v2,pp.1-13,[令和2年7月1日検索],インターネット
Attorney, Agent or Firm:
Yasuhiko Murayama
Shinya Mitsuhiro
Tatsuhiko Abe