Processing

Please wait...

Settings

Settings

1. CN108780521 - ASSOCIATIVE LONG SHORT-TERM MEMORY NEURAL NETWORK LAYERS

Office China
Application Number 201680081162.X
Application Date 30.12.2016
Publication Number 108780521
Publication Date 09.11.2018
Publication Kind A
IPC
G PHYSICS
06
COMPUTING; CALCULATING; COUNTING
N
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3
Computer systems based on biological models
02
using neural network models
04
Architecture, e.g. interconnection topology
G06N 3/04
CPC
G06N 3/08
G06N 3/04
G06N 3/0445
Applicants DEEPMIND TECHNOLOGIES LIMITED
渊慧科技有限公司
Inventors DANIHELKA IVO
伊沃·达尼赫尔卡
KALCHBRENNER NAL EMMERICH
纳尔·埃默里赫·卡尔赫布伦纳
WAYNE GREGORY DUNCAN
格雷戈里·邓肯·韦恩
URIA-MARTINEZ BENIGNO
贝尼尼奥·乌里亚-马丁尼兹
GRAVES ALEXANDER BENJAMIN
亚历山大·本杰明·格拉韦斯
Agents 中原信达知识产权代理有限责任公司 11219
中原信达知识产权代理有限责任公司 11219
Priority Data 62/291,467 04.02.2016 US
Title
(EN) ASSOCIATIVE LONG SHORT-TERM MEMORY NEURAL NETWORK LAYERS
(ZH) 关联长短期记忆神经网络层
Abstract
(EN)
Systems, methods, and apparatus, including computer programs encoded on a computer storage medium, related to associative long short-term memory (LSTM) neural network layers configured to maintain N copies of an internal state for the associative LSTM layer, N being an integer greater than one. In one aspect, a system includes a recurrent neural network including an associative LSTM layer, whereinthe associative LSTM layer is configured to, for each time step, receive a layer input, update each of the N copies of the internal state using the layer input for the time step and a layer output generated by the associative LSTM layer for a preceding time step, and generate a layer output for the time step using the N updated copies of the internal state.

(ZH)
与关联长短期记忆(LSTM)神经网络层有关的系统、方法和设备,包括在计算机存储介质上编码的计算机程序,所述关联LSTM神经网络层被配置成为所述关联LSTM层维护内部状态的N个副本,N是大于一的整数。在一个方面中,一种系统包括:递归神经网络,所述递归神经网络包括关联LSTM层,其中,所述关联LSTM层被配置成,针对每个时间步,接收层输入,使用所述时间步的所述层输入和由所述关联LSTM层针对前一个时间步而生成的层输出来更新所述内部状态的N个副本中的每一个,并且使用所述内部状态的N个更新的副本来生成所述时间步的层输出。

Also published as