Introduction to Machine Learning

Slide 1

Slide 1 text

Tiago Martinho @martinho_t tiagomartinho Introduction to Machine Learning

Slide 2

Slide 2 text

What is ML?

Slide 3

Slide 3 text

Computer science Artiﬁcial Intelligence Machine Learning Pattern Recognition and Computational Learning Theory

Slide 4

Slide 4 text

"the ability to learn without being explicitly programmed” Arthur Samuel,1959

Slide 5

Slide 5 text

"A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P if its performance at tasks in T, as measured by P, improves with experience E.” Tom M. Mitchell

Slide 6

Slide 6 text

Task Detecting Handwriting Characters

Slide 7

Slide 7 text

Task Experience Detecting Handwriting Characters Labelled Handwriting Characters

Slide 8

Slide 8 text

Task Performance Experience Detecting Handwriting Characters Detects Characters w/ Higher Accuracy Labelled Handwriting Characters

Slide 9

Slide 9 text

Why ML?

Slide 10

Slide 10 text

MNIST simple computer vision dataset ML Hello World

Slide 11

Slide 11 text

No content

Slide 12

Slide 12 text

28x28 = 784 numbers

Slide 13

Slide 13 text

uses the examples to automatically infer rules for recognising handwritten digits 0 1 2 3 4 5 6 7 8 9

Slide 14

Slide 14 text

ML Applications

Slide 15

Slide 15 text

Fraud Detection Self-Driving Cars OCR Search engines Computer Vision Health Monitoring … NLP

Slide 16

Slide 16 text

OrCam http://www.orcam.com

Slide 17

Slide 17 text

Alpha Go https://techcrunch.com/2017/05/23/googles-alphago-ai-beats-the-worlds-best-human-go-player/

Slide 18

Slide 18 text

Poker https://www.scientiﬁcamerican.com/article/time-to-fold-humans-poker-playing-ai-beats-pros-at-texas-hold-rsquo-em/

Slide 19

Slide 19 text

How it works

Slide 20

Slide 20 text

Supervised Learning

Slide 21

Slide 21 text

Supervised Learning

Slide 22

Slide 22 text

Supervised Learning General Rule Y = M*x + b

Slide 23

Slide 23 text

Supervised Learning

Slide 24

Slide 24 text

Unsupervised Learning

Slide 25

Slide 25 text

Unsupervised Learning

Slide 26

Slide 26 text

Unsupervised Learning

Slide 27

Slide 27 text

Support Vector Machine

Slide 28

Slide 28 text

SVM

Slide 29

Slide 29 text

Anomaly detection

Slide 30

Slide 30 text

Anomaly detection

Slide 31

Slide 31 text

Anomaly detection

Slide 32

Slide 32 text

Anomaly detection

Slide 33

Slide 33 text

Training Inference

Slide 34

Slide 34 text

Features

Slide 35

Slide 35 text

Collect Train Classify

Slide 36

Slide 36 text

Data 1. Train (60%) 2. Test (20%) 3. Validation (20%) Can we generalise?

Slide 37

Slide 37 text

No content

Slide 38

Slide 38 text

No content

Slide 39

Slide 39 text

No content

Slide 40

Slide 40 text

Tiago Martinho @martinho_t tiagomartinho Thank you!