파이콘 2016에서 발표된 텐서플로우 관련 자료를 정리해 보았습니다.

• Creating AI chat bot with Python 3 and Tensorflow [slide] [pdf]
• Deep Learning with Python & TensorFlow [pdf]
• Introduction to deep learning for machine vision tasks using Keras [pdf]
• 지적 대화를 위한 깊고 넓은 딥러닝 (Feat. TensorFlow) [slide]

특히 ‘지적 대화를 위한 깊고 넓은 딥러닝’을 발표한 김태훈님이 관련 깃허브 레파지토리를 일목요연하게 정리해서 페이스북에 올려 주셨습니다.

1. 이미지(사람의 얼굴 사진)을 이해하고 스스로 만드는 모델
   http://carpedm20.github.io/faces/
   https://github.com/carpedm20/DCGAN-tensorflow
2. 픽셀을 하나씩 예측하며 이미지를 만드는 모델
   https://github.com/carpedm20/pixel-rnn-tensorflow
3. Atari 게임을 화면의 픽셀만 보고 배우는 모델
   https://github.com/devsisters/DQN-tensorflow/
4. 이미지 버전의 ‘왕 – 남자 + 여자 = 여왕’
   https://github.com/carpedm20/visual-analogy-tensorflow
5. 뉴럴 네트워크로 만든 튜링 머신
   https://github.com/carpedm20/NTM-tensorflow
6. 강화 학습 모델들
   https://github.com/carpedm20/deep-rl-tensorflow/
7. Question Answering, Language Model
   https://github.com/carpedm20/MemN2N-tensorflow
8. Character-level Language Models
   https://github.com/carpedm20/lstm-char-cnn-tensorflow
9. Teaching Machines to Read and Comprehend
   https://github.com/carpedm20/attentive-reader-tensorflow
10. Neural Variational Inference for Text Processing
    https://github.com/carpedm20/variational-text-tensorflow
11. Text-based Games using Deep Reinforcement Learning
    https://github.com/carpedm20/text-based-game-rl-tensorflow
12. Continuous Deep Q-Learning with Normalized Advantage Functions
    https://github.com/carpedm20/NAF-tensorflow
13. Asynchronous Methods for Deep Reinforcement Learning
    https://github.com/devsisters/async-rl-tensorflow
14. Neural Abstractive Summarization
    https://github.com/carpedm20/neural-summary-tensorflow