description	A slide for the paper that given below

Report

Sequence to Sequence Learning with Neural Networks

Introduction

• Background-info: Given a large labeled training sets,Deep Neural NetWorks(DNNs) can not be used to map sequences to sequences(STS).
  • The limitation of DNNs: It can only be applied to problems whose inputs and targets can be sensibly encoded with vectors of fixed dimension.
• Objective: A general end-to-end approach to sequence learning that makes minimal assumptions on the sequence structure.
• Method: Long Short-Term Memory(LSTM) can solve general the STS problems.The advantages of LSTM is that it learns to map an input sentence of variable length into a fixed-dimensional vector representation.

Model Structure

• Two different LSTMs: *