Book Image

Clojure for Machine Learning

By : Akhil Wali
Book Image

Clojure for Machine Learning

By: Akhil Wali

Overview of this book

<p>Clojure for Machine Learning is an introduction to machine learning techniques and algorithms. This book demonstrates how you can apply these techniques to real-world problems using the Clojure programming language.</p> <p>It explores many machine learning techniques and also describes how to use Clojure to build machine learning systems. This book starts off by introducing the simple machine learning problems of regression and classification. It also describes how you can implement these machine learning techniques in Clojure. The book also demonstrates several Clojure libraries, which can be useful in solving machine learning problems.</p> <p>Clojure for Machine Learning familiarizes you with several pragmatic machine learning techniques. By the end of this book, you will be fully aware of the Clojure libraries that can be used to solve a given machine learning problem.</p>

Related Content you might be interested in

Current Title:

Small book image
Clojure for Machine Learning
Akhil Wali
Apr, 2014 | 292 pages
No titles found
Table of Contents (17 chapters)
Clojure for Machine Learning
Clojure for Machine Learning
Credits
Credits
About the Author
About the Author
About the Reviewers
About the Reviewers
www.PacktPub.com
www.PacktPub.com
Preface
Preface
Free Chapter
1
Working with Matrices
Working with Matrices
Introducing Leiningen
Representing matrices
Generating matrices
Adding matrices
Multiplying matrices
Transposing and inverting matrices
Interpolating using matrices
Summary
2
Understanding Linear Regression
Understanding Linear Regression
Understanding single-variable linear regression
Understanding gradient descent
Understanding multivariable linear regression
Understanding Ordinary Least Squares
Using linear regression for prediction
Understanding regularization
Summary
3
Categorizing Data
Categorizing Data
Understanding the binary and multiclass classification
Understanding the Bayesian classification
Using the k-nearest neighbors algorithm
Using decision trees
Summary
4
Building Neural Networks
Building Neural Networks
Understanding nonlinear regression
Representing neural networks
Understanding multilayer perceptron ANNs
Understanding the backpropagation algorithm
Understanding recurrent neural networks
Building SOMs
Summary
5
Selecting and Evaluating Data
Selecting and Evaluating Data
Understanding underfitting and overfitting
Varying the regularization parameter
Understanding learning curves
Improving a model
Using cross-validation
Building a spam classifier
Summary
6
Building Support Vector Machines
Building Support Vector Machines
Understanding large margin classification
Linear classification using SVMs
Using kernel SVMs
Summary
7
Clustering Data
Clustering Data
Using K-means clustering
Using hierarchical clustering
Using Expectation-Maximization
Using SOMs
Reducing dimensions in the data
Summary
8
Anomaly Detection and Recommendation
Anomaly Detection and Recommendation
Detecting anomalies
Building recommendation systems
Content-based filtering
Collaborative filtering
Using the Slope One algorithm
Summary
9
Large-scale Machine Learning
Large-scale Machine Learning
Using MapReduce
Querying and storing datasets
Machine learning in the cloud
Summary
References
References
Chapter 1
Chapter 2
Chapter 3
Chapter 4
Chapter 5
Chapter 6
Chapter 7
Chapter 8
Chapter 9
Index
Index

Understanding gradient descent

The gradient descent algorithm is one of the simplest, although not the most efficient techniques to formulate a linear model that has the least possible value for the cost function or error of the model. This algorithm essentially finds the local minimum of the cost function for a formulated linear model.

As we previously described, a three-dimensional plot of the cost function for a single-variable linear regression model would appear as a convex or bowl-shaped surface with a global minimum. By minimum, we mean that the cost function has the least possible value at this point on the surface of the plot. The gradient descent algorithm essentially starts from any point on the surface and performs a sequence of steps to approach the local minimum of the surface.

This process can be imagined as dropping a ball into a valley or between two adjacent hills, as a result of which the ball slowly rolls towards the point that has the least elevation above sea level....

Previous Section
End of Section 3
Next Section