DeepLearning Tutorial

I. Introduction materials

Complete AI learning route, the most detailed collation of Chinese and English resources :star:

AiLearning: MachineLearning – MachineLearning – ML, DeepLearning – DeepLearning – DL, natural language processing NL

Mathematical basis

  • Matrix calculus
  • The mathematical foundations of machine learning
  • CS229 Fundamentals of linear algebra and probability Theory

Fundamentals of machine learning

Quick start

  • Recommended from front to back
  • Machine learning algorithm map
  • Machine Learning by Endar Ng Coursera
  • Chinese Translation && Machine Learning for CS229 By Ng
  • Hundred pages of machine learning

Deep understanding of

  • Recommended from front to back
  • Statistical Learning Methods. Hang Li
  • Pattern Recognition and Machine Learning by Christopher Bishop
  • Machine Learning, Zhihua Zhou
  • PelerHarrington, Machine Learning in action
  • Machine learning and Deep Learning books list

Fundamentals of deep learning

Quick start

  • Recommended from front to back
  • Deep learning Mind Mapping && Deep learning algorithm map
  • Stanford Deep Learning Basics by Andrew Ng
  • Deep learning personal notes && video by Ng
  • MIT Fundamentals of Deep Learning — video Courses 2019
  • Taiwan University (NTU) li Hongyi teaches the course
  • Grokking-Deep-Learning
  • Neural Networks and Deep Learning by Michael Nielsen    
  • CS321-Hinton
  • CS230: Deep Learning
  • CS294-112
Computer vision
  • CS231 Fei Fei Li has authorized personal translation of notes && videos
  • Computer vision
Natural language processing
  • CS224n: Natural Language Processing with Deep Learning
  • NLP hands-on tutorial
  • Introduction to NLP Recommended Books (2019 Edition)
Deep reinforcement learning
  • CS234: Reinforcement Learning

Deep understanding of

  • “Deep Learning” by Yoshua Bengio.Ian GoodFellow: Star:
  • Natural Language Processing by Jacob Eisenstein
  • Reinforcement learning && 2nd edition
  • Hangdong’s deep learning blog, recommended papers
  • Practical Deep Learning for Coders, v3
  • Tensorflow Practical Google Deep Learning Framework by Zheng Zeyu and Gu Siyu

Some books

  • Latest in 2019 – Deep Learning, Generative confrontation, Pytorch excellent textbook recommendations

Engineering capabilities

  • C++ implementation of algorithms in introduction to algorithms
  • Machine learning algorithms in action
  • Deep learning framework
  • How to become an algorithm engineer && From the little white to the entry algorithm, I share my experience with you ~ & my graduate student in these three years :star:
  • AI Algorithm Engineer Manual
  • How to prepare for the interview of algorithm engineer and get the offer of machine learning position in a first-tier Internet company?
  • 【 End 】 Deep learning CV algorithm engineers from the entry to the primary interview how far, about 25 articles
  • Computer related technical interview essentials && Interview algorithm notes – Chinese
  • Interview for Algorithm Engineer
  • Deep learning interview questions
  • Deep learning 500 questions
  • AI algorithm post job search strategy
  • Kaggle of actual combat
    • Common algorithms:
      • Feature Engineering: Continue variable && Categorical Variable
      • Classic Machine Learning Algorithm: LR, KNN, SVM, Random Forest, GBDT(XGBoost&&LightGBM), Factorization Machine, Field-aware Factorization Machine, Neural Network
      • Cross Validation, Model Selection: Grid Search, random Search, hyper-opt
      • Ensemble learning
    • Kaggle Project hands-on (tutorial) = documentation + code + video
    • Kaggle entry series :(1) machine learning environment construction && Kaggle entry series :(2) introduction to Kaggle && Kaggle entry series (3) Titanic first try
    • Go from zero to one into Kaggle
    • Kaggle starter guide
    • A framework to solve Almost all the Machine Learning problems && Approaching (Almost) Any Machine Learning Problem | Abhishek Thakur
    • Take you to the Kaggle Top 1% in minutes
    • How to reach the top 2% of Kaggle competition? Here is a feature exploration experience post
    • How to get into the top 10% in Kaggle’s debut?
  • Big Data & Machine learning related contests recommended

2. Overview of neural network model

  • 1. Understand 25 neural network models
  • 2. DNN Overview Paper: Detailed explanation of feedforward, convolution and cyclic neural network techniques
  • 3. colah’s blog
  • 4. Model Zoom
  • 5. Within DNN overview
  • 6. From basic principles to gradient descent, Xiao Bai can understand the neural network tutorial
  • A collection of machine learning/deep learning overview projects on GitHub

CNN

The history of

  • 1. 94 pages review of convolutional neural networks: From basic Technologies to Research perspectives
  • 2. Lenet-5 to DenseNet
  • 3. A Brief history of CNN Image Segmentation: From R-CNN to Mask R-CNN
  • 4. The Past life of object detection based on deep learning (Mask R-CNN)
  • 5. Overview of lightweight convolutional neural network: SqueezeNet, MobileNet, ShuffleNet, Xception
  • 6. A comprehensive review of deep learning target detection models: Faster R-CNN, R-FCN and SSD
  • 7. Semantic segmentation Survey
  • 7. From RCNN to SSD, this should be the most complete inventory of target detection algorithms
  • 8. Semantic segmentation Survey
  • 9. Overview of semantic segmentation development
  • Deep learning classification network
  • Network Inception, Xception, MobileNet, ShuffeNet, ResNeXt, SqueezeNet, EfficientNet, MixConv
  • The development of CNN network structure
  • Structural Evolution of Convolutional Neural Networks (Form Hubel and Wiesel to SENet)
  • From VGG to NASNet, this paper gives an overview of image classification networks
  • From RCNN to YOLO
  • A Review of Target Detection under the Rule of Various Variants of Faster R-CNN, SSD and YOLO in post-R-CNN Era: Did The Faster R-CNN Series Win?
  • Target detection – Original code summary for 20 models
  • Target detection algorithm overview trilogy
    • A Review of Object Detection Algorithms Based on Deep Learning (I)
    • A Review of Object Detection Algorithms Based on Deep Learning (II)
    • A Review of Object Detection Algorithms Based on Deep Learning (III)
  • How to Approach deep Learning face Recognition? You need this long review | with open source code
  • A review of face detection and recognition algorithms      
    • A review of face detection algorithms          
    • Face detection background and development status
    • Evolution of face recognition algorithms
    • CascadeCNN  
    • MTCNN
    • awesome-Face_Recognition
    • A review of heterogeneous face recognition research
    • The boss is coming: face recognition + mobile push, the boss is coming you immediately know. Face recognition project, network model, loss function, data set summary
    • A Survey of Face Recognition Technology Based on Deep Learning && How to Approach Deep Learning Face Recognition? You need this long overview && Overview of Face Recognition Loss Functions
  • A review of deep learning image super-resolution
  • Evolutionary history of target detection
  • I read all 21 latest papers on target detection (Tencent /Google/ Sensetime/Megvii/Tsinghua/Zhejiang University /CMU/ Huaske/Chinese Academy of Sciences, etc
  • 1. Arxiv2015_baidu_DenseBox, how to evaluate the latest Anchor-Free target detection model FoveaBox? , FCOS: the latest one-stage per-pixel target detection algorithm && the latest Anchor-free target detection model FCOS, now open source! && Huawei Noah proposed CenterNet, one-stage Detector can reach 47AP, has been open source! && AnchorFreeDetection
  • FPN optimization of an overview of target detection algorithms
  • Talk about the “past life” of Anchor (PART 1) && Talk about the “past life” of Anchor (Part 2)
  • [CVPR2019 official Announcement] Pedestrian Re-identification Paper, 2019 Pedestrian Re-identification Annual progress Review
  • Review of 2019CVPR text detection
  • From SRCNN to EDSR, the development process of end-to-end super-resolution methods in deep learning is summarized
  • [CVPR2019 official Announcement] Pedestrian re-recognition paper
  • A review of text detection and recognition techniques in natural scenes
  • Awesome-Image-Colorization
  • Awesome-Edge-Detection-Papers
  • OCR word processing
  • awesome-point-cloud-analysis
  • Graph Neural Network (GNN) review
  • Summary of few-shot Learning
  • Ultra-full deep learning fine-grained image analysis: Projects, reviews, tutorials
  • The decade of image retrieval top and bottom

The tutorial

  • How convolutional neural networks work
  • “The gift of Tanabata” : one day to understand convolutional neural network
  • A Comprehensive Introduction to Different Types of Convolutions in Deep Learning
  • Deformable convolution kernel, separable convolution
  • Understanding of deeply separable convolution, grouping convolution, extended convolution, transposed convolution (deconvolution)
  • All kinds of convolution
  • How many kinds of convolution are there? Understanding convolution in deep learning
  • deconvolution
  • Convolution Network and its variants (deconvolution, extended Convolution, causal Convolution, graph Convolution)
  • How do you evaluate the latest Octave Convolution?
  • Fundamentals of deep learning — convolution types
  • Dilated/Atrous convolution
  • ShuffleNet of CNN model
  • This article describes ResNet and its variants
  • ResNet parsing
  • The first work that introduced CNN into target detection: R-CNN
  • Deep learning (18) Object detection based on R-CNN
  • R-cnn essay details
  • Faster R-CNN Object Detection in Deep Learning
  • First understand the working principle of Mask R-CNN, and then build the color filler application
  • Example Segmentation – Full explanation of Mask RCNN (ROI Align/Loss Function)
  • Semantic segmentation convolutional Neural Network quick introduction
  • The working principle of image semantic segmentation and the change of CNN architecture
  • CapsNet introduction series
    • One of CapsNet’s introductory series: Intuition behind capsule Networks
    • CapsNet Introduction Series 2: How do capsules Work
    • CapsNet starter series 3: Intersac dynamic Routing algorithms
    • CapsNet Starter series 4: Capsule Network Architecture
  • YOLO
  • Target detection | YOLOv2 principle and implementation (YOLOv3 attached)
  • Target detection model YOLO V3 was developed
  • Attention， 1，2，3，4，5
  • Understanding 1×1 convolution kernel in convolutional neural network
  • CornerNet for target detection, 1, 2, 3
  • The performance evaluation index of object detection && NMS and the confidence threshold and IoU threshold of mAP calculation && vernal mAP
  • Crowd count, one, two, three
  • RelationNetwork
  • ShuffleNet V2 and four network architecture design guidelines
  • 【Tensorflow】 tb.nn.depthwise_conv2d how to implement deep convolution?
  • Tensorflow】tf.nn. Atrous_conv2d How to implement empty convolution?
  • 【Tensorflow】tf.nn. Separable_conv2d
  • 【TensorFlow】tf.nn. Conv2d_transpose
  • How to better understand “Focal Loss” of God He Kaiming?
  • How to calculate the number of flops and parameters required for CNN model?

Action

  • First read CapsNet architecture and then use TensorFlow implementation
  • TensorFlow Object Detection API tutorial
    • TensorFlow Object Detection API Tutorial 1
    • TensorFlow Object Detection API Tutorial 2
    • TensorFlow Object Detection API Tutorial 3
    • TensorFlow Object Detection API Tutorial 4
    • TensorFlow Object Detection API Tutorial 5
  • Step by step attention mechanism is implemented using RoI pooling in TensorFlow+Keras environment
  • Mxnet how to view the number of parameters && Mxnet how to view the number of parameters in Mxnet

GAN

  • Su Jianlin blog, explained incisively and vividly

The history of

  • Weird GAN variants
  • The GAN Landscape: Losses, Architectures, Regularization, and indemnification
  • Deep Learning Nova: Fundamentals, Applications, and Trends for GAN
  • A review of GAN generated images

The tutorial

  • 1. GAN principle learning notes
  • 2. Antagonistic generation network of extreme image compression
  • 3. Li Hongyi GAN Course, Taiwan University
    • Basic
    • Improving
  • 4. GAN Computer Vision 2017
  • 5. Implementing GAN on Keras: Building an application that eliminates image blur
  • 6. CycleGAN: image style, want to change change | ICCV paper 2017
  • 7. Wasserstein GAN
  • Use variational inference to understand the generative models (VAE, GAN, AAE, ALI)

Action

  • 1. GAN learning guide: from the principle of entry to the production of Demo
  • 2. GitHub Project of Machine Heart: Complete theoretical derivation and implementation of GAN

RNN

The history of

  • Starting from SRNN in 1990s, this paper reviews the research progress of recurrent neural networks in 27 years

The tutorial

  • Fully illustrated RNN, RNN variants, Seq2Seq, Attention mechanism
  • “Recurrent Neural Networks” (RNN)
  • Introduction and formula combing of RNN and LSTM
  • Deep learning has five – cycle neural networks                      
  • Lossless file compression using recurrent Neural Networks: Stanford University proposes DeepZip
  • Andrew Ng series Modeling course
    • Recurrent Neural Networks (RNN)
    • Coursera: Serial Modeling course Notes (2) — NLP & Word Embeddings
    • Sequence Models & Attention Mechanism (3
  • Word2Vec
    • Word2vec principle (I) CBOW and Skip-Gram model basis
    • Principle of Word2VEC (II) Is based on Hierarchical Softmax model
    • Word2vec principle (iii) Model based on Negative Sampling
    • Learn word2vec with Gensim
  • Talk about the Transformer

Action

  • Tensorflow RNNcell source code analysis and custom RNNcell method
  • The correct opening of the RNN implementation in TensorFlow
  • TensorFlow RNN code
  • Collection of deep NLP models implemented by Tensorflow
  • nlp-tutorial

LSTM

The tutorial

  • 1. Understand long – and short-term memory (LSTM) neural networks
  • 2. Read LSTM and RNN
  • 3. Explore LSTM: basic concepts to internal architecture
  • 4. Translation: In-depth understanding of LSTM series            
    • An in-depth understanding of LSTM Networks (PART 1)
    • An in-depth understanding of LSTM Networks (II)
  • LSTM

Action

  • How to predict stock prices with Tensorflow LSTM
  • Multi-level LSTM practices for TensorFlow
  • Text generation of Anna Karenina — Building LSTM Model using TensorFlow

GNN

The history of

  • Graph Neural Network (GNN) review
  • Graph model in the era of deep learning, Tsinghua published a review of graph network
  • A review of graph neural networks in Tsinghua University: Models and applications
  • Third bullet: GNN overview from IEEE Fellow
  • The most complete GNN literature collation && awesome-graph-neural Networks

The tutorial

  • Graph Convolutional Network (GCN)
  • A brief introduction to Graph Convolutional Networks (GCN)
  • Graph convolutional Networks (GCN) complete guide to novice village

Action

  • What exactly do graph convolutional networks do, which is a minimalist Numpy implementation
  • DGL

Optimization of depth model

  • 1. Overview of optimization algorithms
  • 2. Gradient descent to Adam
  • 3. From gradient descent to Quasi-Newton method: Five learning algorithms for training neural networks
  • 4. Regularization technology summary
    • The most comprehensive review and analysis of regularization techniques ever –part1
    • The most comprehensive review and analysis of regularization techniques ever –part2
  • 5. Optimization Algorithm Series (Math)
  • 6. Gradient disappearance and gradient explosion in neural network training        
  • 7. Optimization and training of neural network
  • 8. Popular explanation of recall rate and precision rate, comprehensive combing: accuracy, accuracy, recall rate, precision rate, recall rate, false positive, true positive,PRC,ROC,AUC,F1
  • 9. Activate function overview
  • Optimizing Deep Neural Networks (3) — Hyperparametric Debugging, Batch Regularization, and Programming Frameworks
  • Machine learning all kinds of entropy
  • 12. Distance and similarity measurement
  • 13. The Black art of Machine learning: Normalization, Regularization
  • 14. Gradient problem of LSTM series
  • 15. Loss function sorting
  • 16. How can residual blocks help solve the gradient dispersion problem
  • 17. FAIR He kaiming et al proposed group normalization: an alternative to batch normalization, which is not limited by batch size
  • 18. Batch Normalization (BN) :1,2,3,4, 5, 6, 7
  • 19. In-depth analysis of exploratory learning, not just BN && How do you distinguish and remember the common types of exploratory algorithms
  • 20. BFGS
  • 21. Explain gradient disappearance, explosion causes and solutions in deep learning in detail
  • 22. Dropout, 1, 2, 3
  • 23. Understanding of Spectral Normalization, common vector Norm and matrix Norm, and Spectral Norm Regularization
  • 24.L1 and L2 regularization
  • 25. Why cross entropy rather than MSE

Four. The alchemist thing

Adjustable and experience

  • Does the trained neural network not work? The penny will get you over the 37 holes
  • Trick neural network training
  • Deep Learning and Computer Vision series (8) neural Network Training and Attention points
  • Neural network training loss does not decrease the cause set
  • Deep learning: Several Solutions to under-fitting problems && over-fitting and under-fitting problems
  • Machine Learning: How to find the Optimal learning rate and Achieve it
  • Unbalanced data set processing methods: first, second, third
  • Whether the expression ability of the same neural network using different activation functions is consistent
  • Overview of gradient descent optimization algorithms, 1, 2
  • Data augmentation of paper notes: Mixup
  • A Guide to avoiding pitfalls: 13 common mistakes made by novice data scientists
  • Why trust CNN’s results? – visual
    • Why should I trust you, my CNN model? (Text 1: CAM and Grad-CAM)
    • Why should I trust you, my CNN model? Snake oil LIME
    • Grad-cam: Visual unity from Deep Networks via gradient-based Localization
    • CV: Keras based target detection using trained HDF5 model to achieve Gradcam of expression or gender in output model
  • Large convolution kernel or small convolution kernel? 1, 2
  • Poor model interpretability? Have you considered the uncertainties?
  • Alchemy Notes series
    • Alchemy Note 1: sample imbalance problem
    • Alchemy Note 2: Data cleansing
    • Alchemy Note 3: Data enhancement
    • Alchemy Note 4: Small sample problems
    • Alchemy Note 5: Data annotation
    • Alchemy Note 6: Blending techniques
    • Alchemy Note 7: Convolutional neural network model design

The strange skill of the ranking

  • The most comprehensive analysis of Kaggle’s six games (PART 1)

  • The most comprehensive analysis of Kaggle six games (2)

Image classification

  • Mentation of Data Augmentation
  • Data Enhancement in Deep Learning (PART 1)
  • Bag of Tricks for Image Classification with CNN && PDF
  • Trick neural network training
  • Kaggle solution sharing
    • Kaggle Image Classification Challenge from 0: Champion solution details
    • The Kaggle Iceberg image classification contest was held recently, to see how the winning team’s proposal is good
    • Image recognition and classification contest, data enhancement and optimization algorithm
    • Identify humpback whale, Kaggle contest first place solution read
    • Kaggle wins gold medal in debut
    • 16-year-old high school student wins Kaggle Landmark Search Challenge! And it was a Kaggle veteran
    • Feelings after 6 Kaggle computer vision competitions
    • Kaggle won third place in its debut – satellite image recognition

Target detection

  • ensemble
  • deformable
  • sync bn
  • ms train/test
  • The optimization strategy of target detection task is tricks
  • Target detection is tricks– sample disequilibrium processing
  • Common TRICK in target detection algorithm
  • Kaggle: Automatic diagnosis system for Lung Cancer 3D Deep Leaky Noisy-or Network
  • Dry goods | great god teach you how to play kaggle – according to CT scans to predict lung cancer

V. Annual summary

  • New Year’s Gift pack: The Heart of Machines 2018 High score tutorial collection
  • Review of CVPR2019 target detection methods

6. Scientific research

Deep learning framework

Python3. X (ap)

  • The Python Tutorial
  • Python tutorial by Liao Xuefeng
  • Novice tutorial    
  • Quick Python tutorial for deep learning beginners – Basics
  • Python – 100 days from Novice to master

Numpy (ap)

  • Quickstart tutorial

  • Numpy Quick Start Numpy 1.14

  • Numpy Chinese document

  • Quick Python tutorial for deep learning beginners – Numpy and Matplotlib

Opencv-python

  • OpenCV-Python Tutorials
  • OpenCV For Python
  • Digital image processing series
  • Python +OpenCV image processing
  • Python Quick Tutorial for deep learning beginners – Python-OpenCV

Pandas

  • Data Science for Python: Pandas

Tensorflow

  • How to learn TensorFlow code efficiently
  • Chinese tutorial
  • TensorFlow official documentation
  • CS20:Tensorflow for DeepLearning Research
  • Ng TensorFlow course
  • The most comprehensive collection of Tensorflow learning resources ever
  • Deep Learning in 21 Projects: A Practical Explanation based on TensorFlow  
  • The most complete Tensorflow2.0 starter tutorial continues to be updated

MXNet

  • Gluon
  • GluonCV
  • GluonNLP

PyTorch

  • Pytorch hands-on deep learning

  • PyTorch Chinese documentation

  • WELCOME TO PYTORCH TUTORIALS

  • The most complete collection of PyTorch learning resources ever

  • The most complete collection of PyTorch learning resources ever

  • Hands-on tour to deep learning with PyTorch

Python visualization

  • Matplotlib Visualizations — The Master Plot (with Full Python Code)
  • Python MatPlotLib tutorial
  • Get started with Matplotlib and start your Python visualization in 10 minutes
  • Quick Python tutorial for deep learning beginners – Numpy and Matplotlib

Annotation tool

  • Target detection annotation tool
    • labelImg
  • Semantic segmentation annotation tool
    • labelme

The data set

  • 1. 25 open data sets related to deep learning
  • 2. Natural Language Processing (NLP) data sets
  • 3) Tang Dynasty poems (43,030)
  • 4. Berkeley makes data sets public
  • 5. ACL 2018 resources: 100+ pre-trained Chinese word vectors
  • 6. Pre-train Chinese word vector
  • 7. Open data set seed banks
  • 8. Computer vision, deep learning, data mining, data set sorting
  • 9. CV Datasets, a well-known computer vision dataset
  • 10. Computer vision related data sets and competitions
  • 11. This is a very comprehensive open source dataset. Are you sure you don’t want it?
  • 12. Characteristics and comparison of existing main data sets for population density estimation
  • 13. DANBOORU2017: A LARGE-SCALE CROWDSOURCED AND TAGGED ANIME ILLUSTRATION DATASET
  • 14. Pedestrian re-identification data set
  • 15. The most complete collation and sharing of common data sets and papers in natural language processing
  • 16. paper, code, sota
  • 17. Megvii RPC large commodity data set release!
  • 18. CVPR 2019 “Near Full Score” Paper: Nvidia Launches first Cross-camera Vehicle Tracking Dataset (Vehicle RE-ID)
  • 19. [OCR technology] Mass generation of text training sets
  • 20. Semantic analysis dataset -MSRA

The meeting list

  • Schedule of international conferences
  • ai-deadlines
  • Keep Up With New Trends
  • Computer conference ranking levels
  • China Computer Society (CCF) recommends international academic journals and conferences

Essay writing tools

  • Windows: Texlive+Texstudio
  • Ubuntu: Texlive+Texmaker

Paper drawing tool

  • Visio2016
  • Matplotlib

Essay Writing Course

  • How to write a qualified NLP paper
  • Liu Yang _ How to write a paper _V7
  • How to Write a Research Paper end to end – Qiu Xipeng
  • Paper Introduction writing one, paper Introduction writing two, paper Introduction writing three
  • How to write the graduation thesis

ResearchGo

  • ResearchGo: Research Life first post — Literature search and management
  • ResearchGo: Research life second post – literature reading
  • ResearchGo: Research Life third post — Reading aid
  • ResearchGo: Research life fourth post — literature research
  • ResearchGo: Research Life fifth post — Literature review
  • ResearchGo: Research life sixth post – How to talk about papers
  • ResearchGo: Research Life 7 – Patent search and application
  • ResearchGo: Research life eighth post – write a paper, do PPT, write documents essential tools collection

Graduation thesis layout

  • Blood vomiting recommended collection of dissertation typesetting tutorial (complete version)
  • How to write a thesis – how to modify the graduation thesis format

Basic theory of machine learning and deep learning

Information theory

  • 1. Entropy in machine learning    
  • 2. Extrapolating KL divergence from Shannon entropy to hand: An overview of information theory in machine learning

Fourier transform

  • Choke on Fourier Analysis tutorial (full version) updated on June 06, 2014
  • How do I summarize the Fourier transform concisely?
  • From the continuous time Fourier series to the Fast Fourier transform
  • Very straightforward FFT (Fast Fourier Transform)
  • Derivation of Fourier series

Deep learning some research areas

Multitasking learning

  • Model Summary -14 Multitask Learning – Overview of Multitask Learning
  • An Overview of multi-task Learning in Deep Neural Networks

Zero Shot Learning

  • Introduction to zero-shot Learning

Small sample Learning (Few Shot Learning)

  • What is fee-shot learning
  • Introduction to zero-shot Learning
  • Summary of few-shot Learning
  • Few-Shot Learning in CVPR 2019
  • When small samples meet machine learning fewshot learning

Multi-view Learning

  • Multi-view Learning introduction
  • Multi-view Learning

Embedding

  • Everything is Embedding, from classical word2vec to deep learning basic operation Item2vec
  • Word Embedding in YJango

word2vec

  • The principle of
    • NLP second understand the essence of word vector Word2vec
    • An easy to understand word2vec
    • Word Embedding in YJango
    • Word vector comparison in NLP: word2vec/glove/fastText/elmo/GPT/Bert
    • Word embedding (word2vec)
    • Talk about how Google word2vec works
    • Why negative sampling in Word2Vec?
  • Training word vector
    • Exercise – word2vec
    • Implementation and application of word2vec method
    • Word2vec, an introduction to Natural Language processing, uses TensorFlow to train word vectors itself
    • Tensorflow is used to train Chinese word vector for Word2VEc
    • How to train word vectors with TensorFlow

The migration study

  • 1. Transfer learning: Classical algorithm analysis
  • 2. What is Transfer Learning? What is the historical future of this field?
  • 3. Transfer study personal notes  
  • One Shot Learning, Zero Shot Learning

Domain adaptive

  • Domain Adaptation video tutorial (with PPT) and classic papers to share
  • 15 Overview of Domain Adaptation, One-shot/zero-shot Learning
  • Deep Transfer Network: Unsupervised Domain Adaptation
  • Unsupervised domain adaptive research based on back propagation
  • Domain adaptation and its application in face recognition
  • CVPR 2018: Object Detection based on Domain adaptation weakly supervised learning

Yuan learning

  • OpenAI proposed a new meta-learning method, EPG, to adjust the loss function to achieve fast training on new tasks      

Reinforcement learning

  • Reinforcement Learning
  • Reinforcement learning materials from entry to abandonment
  • Introduction to reinforcement learning
    • Introduction to Reinforcement learning lecture 1 MDP
  • Reinforcement Learning — What happened from Q-Learning to DQN?
  • From Reinforcement Learning to Deep Reinforcement Learning (I)                  
  • From Reinforcement Learning to Deep Reinforcement Learning (II)
  • This article will help you understand the search strategy of Q-Learning

Recommendation system

  • Documentation related to recommendation algorithms
  • Ten required Papers for Embedding from Entry to Expert
  • The road to recommendation systems
    • Road of recommendation System (1) : Take the road of recommendation system
    • Recommendation System approach (2) : Product clustering

Natural Language Processing (NLP)

  • Word Vector Training Based on Word2VEc
  • Word Vector Training Based on Word2VEC
  • Self-attention Mechanism in Natural Language Processing;    
  • A review of attention mechanisms in natural language processing
  • Word Embedding in YJango
  • CMU& Google Brain proposed a new question answering model QANet

Semantic segmentation correlation algorithm

  • An overview of the dry goods | article mainly semantic network segmentation
  • IU, IoU(Intersection over Union) in deep learning
  • Selective Search for Object Detection (translation)
  • NMS — Non-maximum suppression
  • Bounding Box Regression details

Machine learning theory and practice

  • Machine learning Principles :star:
  • ID3, C4.5, CART, Random forest, Bagging, Boosting, Adaboost, GBDT, XGBoost algorithm Summary
  • Data mining ten algorithms brief introduction, machine learning ten classical algorithms, [algorithm model] easy to understand ten commonly used machine learning algorithms
  • AdaBoost to GBDT series
    • When we are talking about GBDT: From AdaBoost to Gradient Boosting
    • When we are talking about GBDT: Gradient Boosting for classification and regression
    • When we talk about GBDT: Other Ensemble Learning algorithms
  • Integrated study of bagging, stacking, boosting concept to understand

Machine learning Theory

Logistic regression

  • 【 Machine learning interview questions 】 Logical regression

Decision Tree

  • Let’s start with a blind date
  • Get yourself a contact lens
  • CART algorithm and tree pruning in Tree regression basics
  • Three decision tree algorithms based on information Theory (ID3,C4.5,CART)
  • Talk about the decision tree pruning algorithm
  • Machine Learning In action Chapter 9 The Tree is Back
  • The decision tree values ID3 and C4.5 are realized
  • Decision tree value CART implementation

Random forests

  • Introduction and combat of Random Forest

Support Vector Machine (SVM)

  • Popular Introduction to SVM July (This article is the best INTRODUCTION to SVM I have seen)      
  • Machine learning in combat (Python3) learning notes (8) : Support vector machine principles of hand tear linear SVM (SMO training process summary is clear and easy to understand)      
  • Understanding and selection of SVM kernel function
  • Radial Basis Function –RBF                        
  • The SVM kernel function        

PCA

  • Detailed explanation of principal component analysis (PCA) principle
  • Illustrated PCA tutorial
  • Mathematical principle of PCA

SVD

  • Powerful matrix singular value decomposition (SVD) and its applications

  • Singular Value Decomposition (SVD)

  • Detailed explanation and derivation of singular value decomposition (SVD) principle

  • Application of SVD in recommendation system and algorithm derivation

LDA

  • Why does the TEXTBOOK LDA look like this?

Label Propagation Algorithm

  • Label Propagation algorithm and Python implementation
    • The resources

Montagaro search tree

  • Monte Carlo tree search introductory guide

Markov decision

  • Markov Processes: Markov Processes
  • Markov Reward Process of Markov Decision Making;
  • Markov’s Bellman Equation for decision process
  • Markov Decision Process(Markov Decision Process)
  • Optimal Value function and Optimal strategy of Markov decision process

GBDT

  • XGBoost LightGBM war
  • An overview of the similarities and differences between XGBoost, Light GBM and CatBoost  
  • Gradient elevation decision tree
  • GBDT principle and application
  • XGBOOST principle article
  • Xgboost introduction and combat (Principle) && XgBoost introduction and combat (combat adjustment)
  • Xgboost is a kaggle game killer
  • GBDT classification principle and Python implementation
  • GBDT principle and use GBDT to construct a new feature -Python implementation
  • Python+GBDT algorithm combat – prediction to achieve 100% accuracy

Integration (Ensemble)

  • Bagging method and Boosting method for integrated learning method
  • Bagging, Boosting, Stacking && model integration methods is introduced: Bagging, Boosting, and Stacking

EM(Expectation maximization)

  • Everybody knows EM algorithm
  • EM algorithm introduction article                      

Gaussian mixture model (GMM)

  • Mathematical principle and application of Gaussian mixture model and EM algorithm
  • Gaussian mixture model (GMM)

Conditional Random Field (CRF, discriminant model)

  • How to understand conditional random airport easily and happily
  • How to explain the conditional Random Field (CRF) model with easy to understand examples? How is it different from HMM?
  • HMM, MHMM,CRF advantages and disadvantages and differences

TSNE

  • Manifold learning – Dimensionality reduction and visualization of high dimensional data
  • tSNE

Spectral clustering

  • An introduction to Spectral Clustering algorithm
  • Cluster 5– spectrum and spectrum clustering

Outlier detection

  • What are the common “anomaly detection” algorithms in data mining?
  • Overview of outlier detection algorithms

Dimension reduction algorithm

  • Data dimension reduction algorithms – from PCA to LargeVis

Machine learning in action

  • How to choose the best machine learning method for your regression problem?
  • Sklearn: Feature extraction, common models, cross validation
  • Machine Learning Course with Python
  • Python3 machine learning

Some plan

  • Free time to sort out the structure of the whole list, and then collect the series of deep learning and machine learning introduction tutorials, and attached with the code implementation, for a comprehensive and simple start
  • Finish docker_practice this semester