機器學習統計學（影印版）

《機器學習統計學（影印版）》是2018年東南大學出版社出版的圖書，作者是[印] Pratap，Dangeti。

基本介紹

• 書名：機器學習統計學（影印版）
• 作者：[印] Pratap，Dangeti
• 出版社：東南大學出版社
• ISBN：9787564177553

內容簡介

機器學習所涉及的複雜統計學知識困擾了很多開發者。知曉統計學知識可以幫助你為給定的問題構建強壯的機器學習最佳化模型。

《機器學習統計學（影印版）》將教你機器學習所需的實現複雜統計計算的相關內容，可以從中獲得監督學習、無監督學習、強化學習等背後的統計學知識。你將看到討論機器學習相關統計學內容的真實案例並熟悉它們。還能學到用於實現建模、調參、回歸、分類、密度採集、向量處理、矩陣等的相關程式。

學完該書，你會掌握機器學習所需的統計學知識。並且能夠將所學新技能套用於任何行業問題。

圖書目錄

Preface

Chapter 1： Journey from Statistics to Machine Learning

Statistical terminology for model building and validation

Machine learning

Major differences between statistical modeling and machine learning

Steps in machine learning model development and deployment

Statistical fundamentals and terminology for model building andvalidation

Bias versus variance trade-off

Train and test data

Machine learning terminology for model building and validation

Linear regression versus gradient descent

Machine learning losses

When to stop tuning machine learning models

Train， validation， and test data

Cross-validation

Grid search

Machine learning model overview

Summary

Chapter 2： Parallelism of Statistics and Machine Learning

Comparison between regression and machine learning models

Compensating factors in machine learning models

Assumptions of linear regression

Steps applied in linear regression modeling

Example of simple linear regression from first principles

Example of simple linear regression using the wine quality data

Example of multilinear regression - step-by-step methodology of model

building

Backward and forward selection

Machine learning models - ridge and lasso regression

Example of ridge regression machine learning

Example of lasso regression machine learning model

Regularization parameters in linear regression and ridge/lasso regression

Summary

Chapter 3： Logistic Regression Versus Random Forest

Maximum likelihood estimation

Logistic regression - introduction and advantages

Terminology involved in logistic regression

Applying steps in logistic regression modeling

Example of logistic regression using German credit data

Random forest

Example of random forest using German credit data

Grid search on random forest

Variable importance plot

Comparison of logistic regression with random forest

Summary

Chapter 4： Tree-Based Machine Learning Models

Introducing decision tree classifiers

Terminology used in decision trees

Decision tree working methodology from first principles

Comparison between logistic regression and decision trees

Comparison of error components across various styles of models

Remedial actions to push the model towards the ideal region

HR attrition data example

Decision tree classifier

Tuning class weights in decision tree classifier

Bagging classifier

Random forest classifier

Random forest classifier - grid search

AdaBoost classifier

Gradient boosting classifier

Comparison between AdaBoosting versus gradient boosting

Extreme gradient boosting - XGBoost classifier

Ensemble of ensembles - model stacking

Ensemble of ensembles with different types of classifiers

Ensemble of ensembles with bootstrap samples using a single type of

classifier

Summary

Chapter 5： K-Nearest Neighbors and Naive Bayes

K-nearest neighbors

KNN voter example

Curse of dimensionality

Curse of dimensionality with 1D， 2D， and 3D example

KNN classifier with breast cancer Wisconsin data example

Tuning of k-value in KNN classifier

Naive Bayes

Probability fundamentals

Joint probability

Understanding Bayes theorem with conditional probability

Naive Bayes classification

Laplace estimator

Naive Bayes SMS spam classification example

Summary

Chapter 6： Support Vector Machines and Neural Networks

Support vector machines working principles

Maximum margin classifier

Support vector classifier

Support vector machines

Kernel functions

SVM multilabel classifier with letter recognition data example

Maximum margin classifier - linear kernel

Polynomial kernel

RBF kernel

Artificial neural networks -ANN

Activation functions

Forward propagation and backpropagation

Optimization of neural networks

Stochastic gradient descent - SGD

Momentum

Nesterov accelerated gradient - NAG

Adagrad

Adadelta

RMSprop

Adaptive moment estimation - Adam

Limited-memory broyden-fletcher-goldfarb-shanno - L-BFGS

optimization algorithm

Dropout in neural networks

ANN classifier applied on handwritten digits using scikit-learn

Introduction to deep learning

Solving methodology

Deep learning software

Deep neural network classifier applied on handwritten digits using Keras

Summary

Chapter 7： Recommendation Engines

Content-based filtering

Cosine similarity

Collaborative filtering

Advantages of collaborative filtering over content-based filtering

Matrix factorization using the alternating least squares algorithm for

collaborative filtering

Evaluation of recommendation engine model

Hyperparameter selection in recommendation engines using grid search

Recommendation engine application on movie lens data

User-user similarity matrix

Movie-movie similarity matrix

Collaborative filtering using ALS

Grid search on collaborative filtering

Summary

Chapter 8： Unsupervised Learning

K-means clustering

K-means working methodology from first principles

Optimal number of clusters and cluster evaluation

The elbow method

K-means clustering with the iris data example

Principal component analysis - PCA

PCA working methodology from first principles

PCA applied on handwritten digits using scikit-learn

Singular value decomposition - SVD

SVD applied on handwritten digits using scikit-learn

Deep auto encoders

Model building technique using encoder-decoder architecture

Deep auto encoders applied on handwritten digits using Keras

Summary

Chapter 9： Reinforcement Learning

Introduction to reinforcement learning

Comparing supervised， unsupervised， and reinforcement learning in detail

Characteristics of reinforcement learning

Reinforcement learning basics

Category 1 - value based

Category 2 - policy based

Category 3 - actor-critic

Category 4 - model-free

Category 5 - model-based

Fundamental categories in sequential decision making

Markov decision processes and Bellman equations

Dynamic programming

Algorithms to compute optimal policy using dynamic programming

Grid world example using value and policy iteration algorithms with basic Python

Monte Carlo methods

Comparison between dynamic programming and Monte Carlo methods

Key advantages of MC over DP methods

Monte Carlo prediction

The suitability of Monte Carlo prediction on grid-world problems

Modeling Blackjack example of Monte Carlo methods using Python

Temporal difference learning

Comparison between Monte Carlo methods and temporal difference

learning

TD prediction

Driving office example for TD learning

SARSA on-policy TD control

Q-learning - off-policy TD control

Cliff walking example of on-policy and off-policy of TD control

Applications of reinforcement learning with integration of machine

learning and deep learning

Automotive vehicle control - self-driving cars

Google DeepMind's AlphaGo

Robo soccer

Further reading

Summary

Index