MIST101: INTRODUCTION
INTO THE MIST OF MACHINE LEARNING
University of Toronto Machine Intelligence Student Team
MIST101 Outline
University of Toronto Machine Intelligence Student Team
What is Machine Learning
University of Toronto Machine Intelligence Student Team
Traditional Approach
University of Toronto Machine Intelligence Student Team
Drawbacks
University of Toronto Machine Intelligence Student Team
Machine Learning Approach
University of Toronto Machine Intelligence Student Team
Applications
University of Toronto Machine Intelligence Student Team
Types of Learning Tasks
University of Toronto Machine Intelligence Student Team
Machine
Learning
Supervised
Learning
Reinforcement
Learning
Unsupervised
Learning
Supervised Learning
“Dog”
“Cat”
Input
Output
Inputs:
Outputs:
Underlying Function
Single-variable function
Image classifier
(Pixels to Category)
Inputs:
Outputs:
Translation
(Language to Language)
Hello baby!
Salut bébé!
Supervised Learning
Function
“Dog”
Function
“Cat”
Example
Object Detection
(Darknet YOLO)
Supervised Learning
Model
“Dog”
Model
“Cat”
Supervised Learning
Model
Artificial Neural Network
Decision Tree
Graphical
Model
Gaussian
Process
Supervised Learning
Model
Artificial Neural Network
Decision Tree
Graphical
Model
Gaussian
Process
SVM
KNN
Ensembles
Supervised Learning
Model
Artificial Neural Network (ANN)
Supervised Learning
Model
Artificial Neural Network (ANN)
Feedforward Neural Network (FNN)
Convolutional Neural Network (CNN)
Recurrent Neural Network (RNN)
University of Toronto Machine Intelligence Student Team
Supervised Learning
Model
Input
Correct Output
“Cat”
“Dog”
Data Pair
{
Loss
Optimize
Prediction
Types of Learning Tasks
University of Toronto Machine Intelligence Student Team
Machine
Learning
Supervised
Learning
Reinforcement
Learning
Unsupervised
Learning
Label
Supervised Learning
(Outline)
Objective
Learning
Construction
Evaluation
Types of Learning Tasks
University of Toronto Machine Intelligence Student Team
Label
Machine
Learning
Supervised
Learning
Reinforcement
Learning
Unsupervised
Learning
Types of Learning Tasks
University of Toronto Machine Intelligence Student Team
Label
Machine
Learning
Supervised
Learning
Reinforcement
Learning
Unsupervised
Learning
No Label
Unsupervised Learning
Unsupervised
Learning
Clustering
Dimensionality
Reduction
Semi-supervised Learning
Unsupervised Learning
Unsupervised Learning
Unsupervised Learning
Unsupervised Learning
Gulrajani et al. 2017
Unsupervised Learning
Types of Learning Tasks
University of Toronto Machine Intelligence Student Team
Label
Machine
Learning
Supervised
Learning
Reinforcement
Learning
Unsupervised
Learning
No Label
Types of Learning Tasks
University of Toronto Machine Intelligence Student Team
Label
Machine
Learning
Supervised
Learning
Reinforcement
Learning
Unsupervised
Learning
Reward
No Label
“Using chocolates as positive reinforcement for what you consider correct behaviours…”
––– Leonard
Reinforcement Learning
https://en.wikipedia.org/wiki/File:Reinforcement_learning_diagram.svg
Reinforcement Learning
Environment
Policy
Observation &
Reward
Action
Reinforcement Learning
Locations of the stones &
Place a stone
Result of the game
Reinforcement Learning
Policy
Action 1
(33%)
Action 2
(33%)
Action 3
(34%)
Expected
Cumulative
Reward
10
20
30
Reinforcement Learning
Policy
Action 1
(20%)
Action 2
(30%)
Action 3
(50%)
Expected
Cumulative
Reward
10
20
30
Reinforcement Learning
Reinforcement Learning
(Outline)
Coming up...
Thank you!
University of Toronto Machine Intelligence Student Team
References