creativescala / case-study-gradient-descent

A case study on implementing gradient descent for machine learning applications

Case Study: Machine Learning by Gradient Descent

Introduction

This case study looks at gradient descent, and the application of gradient descent to machine learning. We look at gradient descent from a programming, rather than mathematical, perspective. We'll start with a simple example that describes the problem we're trying to solve and how gradient descent can be used to solve it. We'll then look at three methods to compute gradients, the core of the problem:

• numerical differentiation;
• symbolic differentiation; and
• automatic differentiation.

See the site for everything.

case-study-gradient-descent's Issues

Cannot run without changing to project code

The root project will not find a Main class for the run command. Need to switch to project code first. Not sure why. Investigate and streamline.

Text doesn't match code

Code has evolved ahead of the text, particularly the implementation for gradient descent. Sweep through and make sure it's synced up.

Add examples of simplification etc.

The symbolic differentiation section doesn't have enough hand holding.