[DeepLearnig RoadMap

추천 : State of the Art in ML

List

• List_Awesome Deep Learning
• Deep Learning Study : carpedm20
• DeepLearning Gallery : List of awesome Deep Learning Projects
• List of Deep Learning Startups : 딥러닝 스타트업 회사 목록
• List of NLP를 위한 딥러닝 가이드 : Sang-Kil Park 블로그
• List of Awesome Computer Vision
• List of Awesome Deep Vision
• Awesome Deep Learning Resources
• List of Deep Learning's Pioneers : 딥러닝 대가 4인의 SNS정보
• List of Rilles of Machine learning : 구글 연구원인 Martin Zinkevich 작성, [한글 리뷰]
• List of DeepLearning Terms : [한글] 딥러닝 용어 사전
• A Tour Of Machine Learning Algorithms
• List of Deeplearning Tools: TensorFlow, Theano, Caffe, MXNet, Apache, SystemML, BigDL, DistBelief
• List of Reading

• List of Deeplearning: Gunho-Choi 모음

• List of All AI Resources at one place

Paper

• An overview of gradient descent optimization algorithms

• [List of Deep Learning Papers Reading Roadmap]9https://github.com/songrotek/Deep-Learning-Papers-Reading-Roadmap): 논문 읽기 로드맵

• Deep Learning Papers by task

• DeepLearning Papers 모음

• Awesome - Most Cited Deep Learning Papers : terryum 작성
• PathNet : 딥마인드의 새 논문
• CVPR VQA Challenge Workshop논문정리
• Foundational papers
• On the Origin of Deep Learning :
• PathNet: Evolution Channels Gradient Descent in Super Neural Networks
• Highway and Residual Networks learn Unrolled Iterative Estimation: 딥러닝의 정확도가 높은 원인으로 unrolled iterative estimation로 설명한 논문 (기존:각각의 계층이 특정한 추상적인 feature를 대표하며 이를 계층적으로 계산하기 때문이라는 "representation view")

• Neural Machine Translation and Sequence-to-sequence Models: A Tutorial: Sequence-to-sequence모델을 이용한 기계 번역

• Failures of Deep Learning: 딥러닝의 제약(??)

• Efficient Processing of Deep Neural Networks: A Tutorial and Survey: 추천, 튜토리얼
  1. DNN 디자인 요소에 대한 이해
  2. 여러 DNN 하드웨어 벤치마크를 평가할 수 있게 됨.
  3. 다양한 DNN 아키텍쳐와 플랫폼 비교
  4. DNN 디자인을 여러 테크닉과 프로세스 이해
  5. 최신 트렌드와 연구 기회 포착 등

Article/Blog

• Awesome - Most Cited Deep Learning Papers: TerryUm

• 딥러닝을 제대로 이해하기 위해서 필요한 배경지식맵

• 카카오AI리포트] 딥러닝 연구의 현재와 미래 (1/2)

• Brief History of Machine Learning

• Gentle Introduction to the Adam Optimization Algorithm for Deep Learning

• 컨볼루셔널 뉴럴넷 (Convolutional Neural Network) : T-Robotics작성글

• Learning AI if You Suck at Math

  • [Part1]
  • [Part2] :  Practical Projects
  • [Part3] : Building an AI Dream Machine or Budget Friendly Special
  • [Part4] : Tensors Illustrated (with Cats!)
  • [Part5] : Deep Learning and Convolutional Neural Nets in Plain English!

• How to Code and Understand DeepMind’s Neural Stack Machine : Andrew Trask, PhD at University of Oxford

• A Practical Introduction to Deep Learning with Caffe and Python
• GDrive:SNU TF Studiy Group Material
• youtube_Facebook’s advice to students interested in artificial intelligence
• A Complete Guide on Getting Started with Deep Learning in Python
• 한글로 설명이 잘되어있는 라온피플 블로그
• MDP and Reinforcement Learning : 마코브 체인과 강화 학습의 관계설명

동영상

• Neural Networks:

Tutorial (series)

• Deep Learning (Adaptive Computation and Machine Learning series) : An MIT Press book, Ian Goodfellow and Yoshua Bengio and Aaron Courville [GitHub]

• UFLDL Tutorial : 강추_Andrew NG Deep Learning Tutorial

• [여러 모델별 설명]http://colah.github.io/

• Neural Networks and Deep Learning : Neural Networks and Deep Learning, [한글정리]

• A Neural Network in 11 lines of Python

  • Part 1 : low-level 단위 구현으로 알아 보는 뉴럴네트워크

• Deep Learning with PaddlePaddle: 도커를 이용한 인터엑티브 서적

• P119: machine-learning-cheat-sheet

• Applied Deep Learning in Python Mini-Course: machinelearningmastery, 14강좌

Material (ppt, pdf)

• Deep Learning: a Next Step?: ppt, 조경현, 뉴욕대

• Machine Learning & Deep Learning With Spark: ppt50,

Ref Material

• Visual Guide : A Visual Introduction to Machine Learning.

• Deeploearning Workflow: Andrew Ng's recommended machine learning workflow

• Deep Neural Network from scratch: 한 페이지로 정리한 DNN [Code]

Deep Visualization toolbox

• Understanding Neural Networks Through Deep Visualization: [중요] Caffe모델에서 CNN 각 레이어에서 보여 지는 값 시각화 , [github], [Tensorflow 버젼]

  • Deep Visualization Toolbox 의 케라스(Keras) 버전격인 퀴버(Quiver, github), 한글설명
  • https://aiexperiments.withgoogle.com/what-neural-nets-see, [Code]

• Tensorflow PlayGround

• Image Kernels: 각 필터 역할 확인 가능한 사이트

추천 모델들

• Neural Network Architectures: 유명한 모델들 전체 설명
• Deep-Q-Networks : 텐서플로우(TensorFlow)를 이용해 간단한 DQN(Deep-Q-Networks) 에이전트를 만들어 보기
• DeepMind's WaveNet paper
• git_Energy-based generative adversarial network ( EBGAN )
• SyntaxNet : 구글이 자연어 처리(정확히는 자연어 이해)를 수행하는 코드, [설치]

CNN/ILSVRC 우승 모델

• [2012_AlexNet]
• [2013_ZF Net]
• [2014_GoogLeNet]
• [2014_VGGNet]
• 2015_Residual Network (ResNet) : Introduced in 2015 by Microsoft Research (MSR), has made history as the first DNN to surpass human performance in the complex task of image classification.
  • ResNet in TensorFlow
  • Deep Residual Networks

• MXNet : Image Classification of ImageNet database

• MNIST Competition Tensorflow KR Group

Datasets

• 얼굴 이미지 : PubFig

• Open Data

• Awesome Public Datasets

THE NEURAL NETWORK ZOO