ProClaim:
之前一直在做CNN的一些研究,最近刚刚回到实验室,定下来了自己的小组,然后开始了一些LSTM的学习。
将近学习了两天半吧,结构弄得差不多了,Theano上LSTM tutorial 的例程也跑了跑,正在读代码ing。
这篇博客主要是我之后要做的一个小报告的梗概,梳理了一下LSTM的特点和适用性问题。
发在这里权当做开博客压压惊。
希望之后能跟各位朋友多多交流,共同进步。
1. 概念:
Long short-termmemory(LSTM)is arecurrent neuralnetwork(RNN)architecture (anartificialneural network)published[1]in 1997 bySepp HochreiterandJürgen Schmidhuber. Like most RNNs, an LSTM network is universalin the sense that given enough network units it can compute anything aconventional computer can compute, provided it has the properweightmatrix, which may be viewed as its program. Unliketraditional RNNs, an LSTM network is well-suited to learn from experience toclassify,processand