Machine Learning

Subtitle

Introduction

 

What is Machine Learning?

A new approach to Artificial Intelligence with an emphasis on statistical analysis, the ability for the computer to write its own set of rules, rather than humans writing all of the conditional logic.

Its purpose

Using the power of computers to complement and supplement human intelligence.

 

Example: programs can scan and process huge databases detecting patterns that are beyond the scope of human perception.

Examples

Face/voice detection on your phone

 

Detecting eye deceases

 

Self-driving cars

Steps involved

Defining a problem

Preparing data

Evaluating algorithms

Improving results

Presenting results

Frameworks

Scikit-learn

TensorFlow

Machine learning on AWS

IBM Watson

Concepts of Learning

 

Categories of Learning

  • Supervised Learning
     
  • Unsupervised Learning
     
  • Semi-supervised Learning

Categories of algorithms

  • Supervised Learning Algorithm
     
  • Unsupervised Learning Algorithm
     
  • Semi-supervised Learning Algorithm
     
  • Reinforcement Learning Algorithm

Supervised Learning

  • Can be classified into 2 types:
    • Regression
    • Classification
  • When a machine intelligence predicts a category or a quantity using models of classification and regression respectively.

Unsupervised Learning

Text

Semi-supervised Learning

Text

Reinforcement Learning

Text

Techniques

 

Techniques used

  • Classification
  • Regression
  • Recommendation
  • Clustering
  •  

Training and testing data

 

Training

  • S

Testing

  • S
Made with Slides.com