List

• List of Awesome Recurrent Neural Networks: Jiwon Kim

• List of NLP를 위한 딥러닝 가이드 : Sang-Kil Park 블로그글

Article / Post

• 순환 신경망 레이어 및 기본 모델 이야기

• (한글 번역) Anyone Can Learn To Code an LSTM-RNN in Python (Part 1: RNN)

• ppt:홍콩과기대 김성훈 교수

• 초보자를 위한 RNNs과 LSTM 가이드: Deeplearning4J 번역 자료

  • DL4J와 RNNs (Recurrent Neural Networks)

  • Understanding LSTM and its diagrams

• Building Jarvis AI with Natural Language Processing : By Mark Zuckerburg, CEO at Facebook.

• Deep Learning Research Review: Natural Language Processing

• Understanding LSTM Networks: colah's Blog 원문, 번역1, 번역2

• ppt: Introduction For seq2seq(sequence to sequence) and RNN

• 4 APPROACHES TO NATURAL LANGUAGE PROCESSING & UNDERSTANDING: LP의 4가지 다른 접근방법

• Distributional : 최근 유행하는 ML이 여기죠~ 폭은 넓힐 수 있지만, 깊이는 잡지 못함
• Frame-based: 마빈 민스키 ... 논리적 semantics 에 강점. 확고한 supervision이 존재해야 한다는 큰 단점이~
• Model-theoretical: Q/A와 rich semantics의 장점. labor-intensive and narrow in scope (프레임 기반보다 더함)
• Interactive learning: language as a cooperative game between speaker and listener ... 이게 앞으로 제일 promising 하지 않을까? ``` 이 개념이 재미있네요.
• Syntax – what is grammatical? : “no compiler errors”
• Semantics – what is the meaning?: “no implementation bugs”
• Pragmatics – what is the purpose or goal?: “implemented the right algorithm.” ... 사실 목표는 여기! ```

• The Unreasonable Effectiveness of Recurrent Neural Networks: RNN의 선능

• RNN Regularizations: RNN 오버피팅 문제 해결법 정리

• RECURRENT NEURAL NETWORKS TUTORIAL

  • PART 1: INTRODUCTION TO RNNS [번역]
  • PART 2: IMPLEMENTING A RNN WITH PYTHON, NUMPY AND THEANO [번역]
  • PART 3: BACKPROPAGATION THROUGH TIME AND VANISHING GRADIENTS [번역]
  • PART 4: IMPLEMENTING A GRU/LSTM RNN WITH PYTHON AND THEANO [번역]

• Unfolding RNNs By Suriyadeepan Ram

  • PART1: Concepts and Architectures
  • PART2: Vanilla, GRU, LSTM RNNs from scratch in Tensorflow

• Building Your Own Neural Machine Translation System in TensorFlow.

• 신경망번역을 위한 seq2seq 영문 튜토리얼

Tutorial

• 신경망번역을 위한 seq2seq 영문 튜토리얼

• 음성인식 관련 텐서플로우 RNN Tutorial: TensorFlow RNN Tutorial, 2017.03.23

• 모두의 연구소:RNN과 LSTM - 쫄지말자 딥러닝

• University of Oxford2017

Implemetation

• Anyone Can Learn To Code an LSTM-RNN in Python

• 모두를 위한 딥러닝 RNN자료: lab-12-1-hello-rnn.py

• Hello_LSTM

• hello_sequence.py: simple Sequence-to-sequence model with an attention mechanism

Papers

• Neural Machine Translation and Sequence-to-sequence Models: A Tutorial