LSTM与Highway-LSTM算法实现的研究概述
[email protected]
http://www.cnblogs.com/swje/
作者:Zhouwan
2015-12-22
声明:
1)该LSTM的学习系列是整理自网上很多大牛和机器学习专家所无私奉献的资料的。具体引用的资料请看参考文献。具体的版本声明也参考原文献。
2)本文仅供学术交流,非商用。所以每一部分具体的参考资料并没有详细对应。如果某部分不小心侵犯了大家的利益,还望海涵,并联系博主删除。
3)本人才疏学浅,整理总结的时候难免出错,还望各位前辈不吝指正,谢谢。
部分参考文献:
[High-way-lstm-1]High-way LSTM RNNS FOR Distant Speech Recognition.pdf : 访问密码 510b
[High-way-lstm-2]Character-Aware Neural Language Models.pdf :访问密码 510b
神经网络课件:NN-Lectures:CSC321 Winter 2015 Introduction to Neural Networks:访问密码 0442
三、我的学习计划如下:
The Unreasonable Effectiveness of Recurrent Neural Networks
(2) Image Captioning
[CVPR15]]Long-term Recurrent Convolutional Networks for Visual Recognition and Description
Deep Visual-Semantic Alignments for Generating Image Descriptions
(3)Speech Recognition
(4)Machine Translation
[NIPS15]Sequence to Sequence Learning with Neural Networks.
[High-way-lstm-1]High-way LSTM RNNS FOR Distant Speech Recognition.pdf : 访问密码 510b
[High-way-lstm-2]Character-Aware Neural Language Models.pdf :访问密码 510b