Machine Learning Basics

Slide 1

Slide 1 text

I ❤ Machine Learning #ML #KillSarahConnor Machine Learning Basics PHPNW, September 2017 @tom_walder

Slide 2

Slide 2 text

Tom Walder @tom_walder CTO at Gear4music.com Manchester, UK Developer at heart github.com/tomwalder

Slide 3

Slide 3 text

Machine Learning Google Trends (last 5 years)

Slide 4

Slide 4 text

Artiﬁcial Intelligence Intelligence exhibited by machines, rather than humans or other animals Wikipedia Machine Learning Giving computers the ability to learn without being explicitly programmed Arthur Samuel (via Wikipedia)

Slide 5

Slide 5 text

Slide 6

Slide 6 text

Applications of ML • Google search

Slide 7

Slide 7 text

Applications of ML • Google search • Facial recognition

Slide 8

Slide 8 text

Applications of ML • Google search • Facial recognition • Apps (translation)

Slide 9

Slide 9 text

Applications of ML • Google search • Facial recognition • Apps (translation) • Self-driving cars

Slide 10

Slide 10 text

Applications of ML • Google search • Facial recognition • Apps (translation) • Self-driving cars • Terminators?

Slide 11

Slide 11 text

A Learning Computer? How do you teach a computer? How can it learn and improve? MATHS + DATA Software! ______

Slide 12

Slide 12 text

Types of ML Algorithm Supervised Know outcomes in data • Classiﬁcation Diseased/Healthy Stop/Go • Regression Height: 186cm Share price: 50p Unsupervised Discovery • Clustering Discover inherent groups • Association People who buy X also buy Y

Slide 13

Slide 13 text

ML Lifecycle Overview Data Learning Prediction Choose & train a mathematical model. Iterative, guided process. Source & prepare your data Executing your model on new, unseen data.

Slide 14

Slide 14 text

Data Data Learning Prediction Identify features that are important to your problem Data format - text, images, video, audio Source, understand, clean, ﬁlter, normalise Split into training & evaluation data The quality of your ﬁnal model will depend on your data.

Slide 15

Slide 15 text

Data prep The quality of your ﬁnal model will depend on your data.

Slide 16

Slide 16 text

My Problem The Perfect Pint

Slide 17

Slide 17 text

Head (mm) Temp (C) Good pint? 14 3.5 Yes 18 14 No 28 5 No 16 4 Yes 13 5 Yes 19 8.2 No 50 0 Not sure Data

Slide 18

Slide 18 text

Learning Data Learning Prediction Select & train a mathematical model Iterative, guided process Computationally intensive Human input is key Cloud, training at scale, GPUs, TPUs Output is a set of algorithms, weights Iterative performance evaluation.

Slide 19

Slide 19 text

Linear Classiﬁer

Slide 20

Slide 20 text

Prediction Data Learning Prediction Executing our trained model on new, unseen data. Online / on-device / anywhere.

Slide 21

Slide 21 text

Libraries for Machine Learning Speciﬁc problems, experimentation

Slide 22

Slide 22 text

github.com/php-ai/php-ml Machine Learning Library for PHP

Slide 23

Slide 23 text

Most Popular ML Library • Python • Originally Google • Open Source in 2015 • Bas become the most widely adopted ML lib • Large range of tools, visualisations • Works seamlessly with cloud- based training and serving products from Google • Runs on devices

Slide 24

Slide 24 text

Pre-trained ML REST APIs Someone else has done the hard work

Slide 25

Slide 25 text

Google ML REST APIs • Natural Language • Speech • Translation • Vision • Video Intelligence Pre-trained machine learning REST APIs for image, text, video and voice analysis.

Slide 26

Slide 26 text

The End • Machine Learning Basics • Tom Walder / @tom_walder • Feedback / https://joind.in/talk/58f6b • Resources • https://github.com/php-ai/php-ml • https://www.tensorﬂow.org/ • https://cloud.google.com/products/machine-learning/ • https://cloud.google.com/dataprep/ Bye