Learning nonregular languages: A comparison of simple recurrent networks and LSTM

被引：47

作者：

Schmidhuber, J ^{[1
]}

Gers, F ^{[1
]}

Eck, D ^{[1
]}

机构：

[1] Ist Dalle Molle Studi Intelligenza Artificiale, CH-6928 Manno Lugano, Switzerland

来源：

NEURAL COMPUTATION | 2002年 / 14卷 / 09期

关键词：

D O I：

10.1162/089976602320263980

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

In response to Rodriguez's recent article (2001), we compare the performance of simple recurrent nets and long short-term memory recurrent nets on context-free and context-sensitive languages.

引用

页码：2039 / 2041

页数：3

共 8 条

[1] [Anonymous], 1997, Neural Comput
[2] Context-free and context-sensitive dynamics in recurrent neural networks
Bodén, M
Wiles, J
[J]. CONNECTION SCIENCE, 2000, 12 (3-4) : 197 - 210
[3] FINDING STRUCTURE IN TIME
ELMAN, JL
[J]. COGNITIVE SCIENCE, 1990, 14 (02) : 179 - 211
[4] LSTM recurrent networks learn simple context-free and context-sensitive languages
Gers, FA
Schtmidhuber, J
[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 2001, 12 (06): : 1333 - 1340
[5] GERS FA, 2001, P ICANNGA 2001 C NEW, V1, P134
[6] Hochreiter S., 2001, FIELD GUIDE DYNAMICA, DOI 10.1109/9780470544037.ch14.
[7] Simple recurrent networks learn context-free and context-sensitive languages by counting
Rodriguez, P
[J]. NEURAL COMPUTATION, 2001, 13 (09) : 2093 - 2118
[8] Sepp H., 1991, Untersuchungen zu dynamischen neuronalen Netzen

← 1 →