S E E D / / S E A R C H F O R E X T R A O R D I N A R Y E X - - PowerPoint PPT Presentation

▶
s e e d s e a r c h f o r e x t r a o r d i n a r y e x p
SMART_READER_LITE
LIVE PREVIEW

S E E D / / S E A R C H F O R E X T R A O R D I N A R Y E X - - PowerPoint PPT Presentation

Sep 10, 2022 350 likes •1.01k views

S E E D / / S E A R C H F O R E X T R A O R D I N A R Y E X P E R I E N C E S D I V I S I O N Deep Learning in Games Martin Singh-Blom @singhblom S E E D // Introduction What is Machine Learning? S E E D // Introduction What is

slide-1
SLIDE 1 S E E D / / S E A R C H F O R E X T R A O R D I N A R Y E X P E R I E N C E S D I V I S I O N
slide-2
SLIDE 2

Deep Learning in Games

Martin Singh-Blom @singhblom

slide-3
SLIDE 3

What is Machine Learning?

S E E D // Introduction

slide-4
SLIDE 4

S E E D // Introduction

What is Artificial Intelligence?

slide-5
SLIDE 5
  • r ...

S E E D // Introduction

slide-6
SLIDE 6

Guess the function!

S E E D // Introduction

slide-7
SLIDE 7

f (2) = 4

S E E D // Introduction

slide-8
SLIDE 8

f (8) = 16

S E E D // Introduction

slide-9
SLIDE 9

f (x) = 2x

S E E D // Introduction

slide-10
SLIDE 10

How does the machine guess?

S E E D // Middle

slide-11
SLIDE 11

It learns from the data.

S E E D // Middle

slide-12
SLIDE 12

S E E D // Middle

(That’s why we call it machine learning!)

It learns from the data.

slide-13
SLIDE 13

S E E D // How do the machines learn?

x y

Guess a straight line!

slide-14
SLIDE 14

S E E D // How do the machines learn?

x y

Guess a straight line! f (x) = 9.5 – 0.3x

slide-15
SLIDE 15

S E E D // How do the machines learn?

x y

Guess a straight line! f (x) = 9.5 – 0.3x

slide-16
SLIDE 16

S E E D // How do the machines learn?

x y

Guess a straight line! f (x) = 0.3 + 0.6x

slide-17
SLIDE 17

S E E D // How do the machines learn?

x y

Guess a straight line! f (x) = 0.3 + 0.6x

slide-18
SLIDE 18

S E E D // How do the machines learn?

Guess a straight line! f (x) = 0.3 + 0.6x f (x) = 9.5 – 0.3x

slide-19
SLIDE 19

S E E D // How do the machines learn?

1. Data – f ( x ) = y pairs. 2. A way to tell the machine how bad a guess is. 3. Some idea of what kind of function the machine is allowed to guess – straight line? Curve? Something stranger?

That is all there is to it!

slide-20
SLIDE 20

S E E D // How do the machines learn?

x y

Guess a straight line!

slide-21
SLIDE 21

S E E D // How do the machines learn?

Guess a straight line!

slide-22
SLIDE 22

What is Deep Learning?

S E E D // Deep Learning

slide-23
SLIDE 23

What are Artificial Neural Networks?

S E E D // Deep Learning

slide-24
SLIDE 24

f

S E E D // Deep learning

slide-25
SLIDE 25

S E E D // Deep Learning

slide-26
SLIDE 26

S E E D // Deep learning

slide-27
SLIDE 27

f ( ) = 8

S E E D // Deep learning

slide-28
SLIDE 28

f ( ) = 8

S E E D // Deep learning

slide-29
SLIDE 29

f ( ) = 5

S E E D // Deep learning

slide-30
SLIDE 30

f ( ) = 0

S E E D // Deep learning

slide-31
SLIDE 31

f ( ) = 6

S E E D // Deep learning

slide-32
SLIDE 32

S E E D // Deep learning

slide-33
SLIDE 33

S E E D // Deep learning

slide-34
SLIDE 34

f ( )

S E E D // Deep learning

slide-35
SLIDE 35

f ( ) = cat

S E E D // Deep learning

slide-36
SLIDE 36

”I saw it in a theater once and it was great. It was very… I don’t know, a little dark. I like the psychological effects and the way it portrays the characters.”

f ( ) =

”Have you seen Suicide Squad?”

S E E D // Deep learning

slide-37
SLIDE 37

f ( ) =

”A person flying a kite

  • n a beach”

S E E D // Deep learning

slide-38
SLIDE 38

f ( ) =

”A coffee, please.”

S E E D // Deep learning

slide-39
SLIDE 39

f ( ) =

”A coffee, please.”

S E E D // Deep learning

slide-40
SLIDE 40

f ( ) =

S E E D // Deep learning

slide-41
SLIDE 41

Agents in Games

S E E D // Agents in Games

slide-42
SLIDE 42

S E E D // Agents in Games

slide-43
SLIDE 43

S E E D // Agents in Games

slide-44
SLIDE 44

S E E D // Agents in Games

f ( ) =

slide-45
SLIDE 45
slide-46
SLIDE 46

S E E D // TOPIC

slide-47
SLIDE 47
slide-48
SLIDE 48

S E E D // Deep learning

slide-49
SLIDE 49

AlphaGo

slide-50
SLIDE 50

Animation

S E E D // Animation

slide-51
SLIDE 51

S E E D // Animation

slide-52
SLIDE 52

S E E D // Animation

f ( ) =

slide-53
SLIDE 53

Audio-Driven Facial Animation by Joint End-to-End Learning of Pose and Emotion, Karras et al., 2017, NVIDIA

slide-54
SLIDE 54

Learn all the things!

S E E D // All the things!

slide-55
SLIDE 55

S E E D // All the things!

f ( ) =

slide-56
SLIDE 56

Physics

Physics Forests: Real-time Fluid Simulation using Machine Learning, Ladicky et al., 2015, www.physicsforests.com

slide-57
SLIDE 57

S E E D // All the things!

slide-58
SLIDE 58

S E E D // All the things!

f ( ) =

slide-59
SLIDE 59

Realtime Multi-Person 2D Human Pose Estimation using Part Affinity Fields, Cao et al., 2017

slide-60
SLIDE 60

S E E D // All the things!

f ( ) =

slide-61
SLIDE 61

Phase-Functioned Neural Networks for Character Control, Holden, 2017

slide-62
SLIDE 62

S E E D // All the things!

f ( ) = f ( ) = 8 f ( ) = f ( ) = f ( ) = f ( ) = cat f ( ) =

”I saw it in a theater once and it was great. It was very… I don’t know, a little dark. I like the psychological effects and the way it portrays the characters.” ”Have you seen Suicide Squad?”

f ( ) = ”A coffee, please.” f ( ) =

”A coffee, please.”
slide-63
SLIDE 63

S E E D // Final remark

Instead of programming – showing Same method for every problem Greatest paradigm change in computing since transistors It’s all just function guessing – or – A new paradigm for computing

slide-64
SLIDE 64

FIN

slide-65
SLIDE 65

S E E D // Thank you

Stockholm

Hector Anadon Leon Jorge del Val Santos Mattias Teye Anastasia Opara Camilo Gordillo Joakim Bergdahl Jack Harmer Linus Gisslén Henrik Johansson Paul Greveson Niklas Nummelin Ken Brown Mark Kyobe Effeli Holst Jenna Frisk Ida Winterhaven Tomasz Stachowiak Colin Barré-Brisebois Graham Wihlidal Lars Sjöström Daniel Lundin

Montreal

Mathieu Lamarre Etienne Danvoye

Los Angeles

Carlos Ochoa JP Lewis Binh Le Henrik Halen John Courte

Special thank you to

Magnus Nordin Johan Andersson