Machine LEARNING 101

 

 

GOAL OF THIS TALK

DEEP LEARNING HAS MASTERED GO

DEEP LEARNING CAN PLAY VIDEO GAMES

DEEP LEARNING CAN PLAY VIDEO GAMES

DEEP LEARNING CAN IMITATE STYLE

EVERYONE IS USING DEEP LEARNING

History?

MILESTONES

THE BREAKTHROUGH (2012)

In 2011, a misclassification rate of 25% was near state of the art on ILSVRC

 

In 2012, Geoff Hinton and two graduate students, Alex Krizhevsky and Ilya Sutskever, entered ILSVRC with one of the first deep neural networks trained on  GPUs, now known as "Alexnet"

 

Result: An error rate of 16%, nearly half what the second place entry was able to achieve.

The computer vision world immediately took notice

WHAT'S CHANGED? WHY NOW?

Several key advancements have enabled the modern deep learning revolution

Massive Data

WHAT'S CHANGED? WHY NOW?

Several key advancements have enabled the modern deep learning revolution

GPUs/CPUs

 

WHAT'S CHANGED? WHY NOW?

Several key advancements have enabled the modern deep learning revolution

Methodological advancements have made deeper networks easier to train

 

Architecture

Optimizers

Activation Functions

WHAT'S CHANGED? WHY NOW?

Several key advancements have enabled the modern deep learning revolution

Open source framework

Give you Feature or not?

Scikit learn -> Classic AI

Tensorflow -> Deep Learning

Demo

MNIST

How it works

Data

Model

Verify

Training

Introduction to Deep Learning Models

Idea is coming from Biology

MULTILAYER PERCEPTRONS (MLPs)

  • Most generic form of a neural net is the "multilayer perceptron"
  • Input undergoes a series of nonlinear transformation before a final classification layer

Company

Billy

Ai

Manager A

manager B

Rugu

Siva

CEO

0.8

1.2

0.7

1.1

2

0.8

0.8

1.1

1.1

0.9

3

5

Company

Billy

Ai

Manager A

Manager B

Rugu

Siva

CEO

0.8

1.2

0.7

1.1

2

0.8

0.8

1.1

1.1

0.9

3

2

(2*0.8)+(1.2*0.8) >

3 ?

(0.8*0.7)+(1.1*1.1) > 2 ?

(2.56*1.1)+(1.77*0.9) > 4 ?Yes or No

4

Is This A or B

Is This Weird?

How Much?

How Many?

How is this organised?

What should I do now?

My Impression

1+1 = 2

Traditional Computer

My Impression

1+1 = 2

AI Computer

1+3 = 4

2+3 = 5

:

10+10 = 20

My Impression

So they provided different way of how to use computer

Reality

Reality

https://www.microsoft.com/en-us/research/wp-content/uploads/2017/03/main.pdf

Deep Learning 101

By Beavis28

Deep Learning 101

  • 282