Mastering Machine Learning Algorithms

Book Image

Mastering Machine Learning Algorithms

Book Image

Mastering Machine Learning Algorithms

Overview of this book

Machine learning is a subset of AI that aims to make modern-day computer systems smarter and more intelligent. The real power of machine learning resides in its algorithms, which make even the most difficult things capable of being handled by machines. However, with the advancement in the technology and requirements of data, machines will have to be smarter than they are today to meet the overwhelming data needs; mastering these algorithms and using them optimally is the need of the hour. Mastering Machine Learning Algorithms is your complete guide to quickly getting to grips with popular machine learning algorithms. You will be introduced to the most widely used algorithms in supervised, unsupervised, and semi-supervised machine learning, and will learn how to use them in the best possible manner. Ranging from Bayesian models to the MCMC algorithm to Hidden Markov models, this book will teach you how to extract features from your dataset and perform dimensionality reduction by making use of Python-based libraries such as scikit-learn v0.19.1. You will also learn how to use Keras and TensorFlow 1.x to train effective neural networks. If you are looking for a single resource to study, implement, and solve end-to-end machine learning problems and use-cases, this is the book you need.

Title Page

Dedication

Packt Upsell

Contributors

Preface

Free Chapter

Machine Learning Model Fundamentals

Machine Learning Model Fundamentals

Models and data

Features of a machine learning model

Loss and cost functions

Introduction to Semi-Supervised Learning

Introduction to Semi-Supervised Learning

Semi-supervised scenario

Generative Gaussian mixtures

Contrastive pessimistic likelihood estimation

Semi-supervised Support Vector Machines (S3VM)

Transductive Support Vector Machines (TSVM)

Graph-Based Semi-Supervised Learning

Graph-Based Semi-Supervised Learning

Label propagation

Label spreading

Label propagation based on Markov random walks

Manifold learning

Bayesian Networks and Hidden Markov Models

Bayesian Networks and Hidden Markov Models

Conditional probabilities and Bayes' theorem

Bayesian networks

Hidden Markov Models (HMMs)

EM Algorithm and Applications

EM Algorithm and Applications

MLE and MAP learning

Gaussian mixture

Factor analysis

Principal Component Analysis

Independent component analysis

Addendum to HMMs

Hebbian Learning and Self-Organizing Maps

Hebbian Learning and Self-Organizing Maps

Sanger's network

Rubner-Tavan's network

Self-organizing maps

Clustering Algorithms

Clustering Algorithms

k-Nearest Neighbors

Spectral clustering

Ensemble Learning

Ensemble Learning

Ensemble learning fundamentals

Gradient boosting

Ensembles of voting classifiers

Ensemble learning as model selection

Neural Networks for Machine Learning

Neural Networks for Machine Learning

The basic artificial neuron

Multilayer perceptrons

Optimization algorithms

Regularization and dropout

Batch normalization

Advanced Neural Models

Advanced Neural Models

Deep convolutional networks

Recurrent networks

Transfer learning

Autoencoders

Variational autoencoders

Generative Adversarial Networks

Generative Adversarial Networks

Adversarial training

Wasserstein GAN (WGAN)

Deep Belief Networks

Deep Belief Networks

Introduction to Reinforcement Learning

Introduction to Reinforcement Learning

Reinforcement Learning fundamentals

Policy iteration

Value iteration

TD(0) algorithm

Advanced Policy Estimation Algorithms

Advanced Policy Estimation Algorithms

TD(λ) algorithm

SARSA algorithm

Other Books You May Enjoy

Other Books You May Enjoy

Leave a review - let other readers know what you think

Index

Customer Reviews

5 star

0

4 star

0

3 star

0

2 star

0

1 star

0

Summary

In this chapter, we presented some fundamental clustering algorithms. We started with KNN, which is an instance-based method that restructures the dataset to find the most similar samples given a query point. We discussed three approaches: a naive one, which is also the most expensive in terms of computational complexity, and two strategies based respectively on the construction of a KD Tree and a Ball Tree. These two data structures can dramatically improve performance even when the number of samples is very large.

The next topic was a classic algorithm: K-means, which is a symmetric partitioning strategy, comparable to a Gaussian mixture with variances close to zero, that can solve many real-life problems. We discussed both a vanilla algorithm, which wasn't able to find a valid sub-optimal solution, and an optimized initialization method, called K-means++, which was able to speed up the convergence towards solutions quite close to the global minimum. In the same section, we also...