Behaviour and Reasoning Description Language (BRDL) Antonio Cerone - - PowerPoint PPT Presentation

behaviour and reasoning description language brdl
SMART_READER_LITE
LIVE PREVIEW

Behaviour and Reasoning Description Language (BRDL) Antonio Cerone - - PowerPoint PPT Presentation

Behaviour and Reasoning Description Language (BRDL) Antonio Cerone Department of Computer Science School of Science and Technology Nazarbayev University Nur-Sultan, Kazakhstan email: antonio.cerone@nu.edu.kz A. Cerone, Nazarbayev University


slide-1
SLIDE 1

Behaviour and Reasoning Description Language (BRDL)

Antonio Cerone Department of Computer Science School of Science and Technology Nazarbayev University Nur-Sultan, Kazakhstan email: antonio.cerone@nu.edu.kz

  • A. Cerone, Nazarbayev University – p.1/26
slide-2
SLIDE 2

BRDL

CIFMA 2019, 17 September 2019

Behaviour and Reasoning Description Language

  • knolwledge representation
  • knowledge retrieval
  • reasoning
  • human behaviour
  • problem solving
  • A. Cerone, Nazarbayev University – p.2/26
slide-3
SLIDE 3

Human Memory

CIFMA 2019, 17 September 2019

Environment

Input Channel

Output Channel

Human Memory

  • A. Cerone, Nazarbayev University – p.3/26
slide-4
SLIDE 4

Human Memory

CIFMA 2019, 17 September 2019

Environment

Input Channel

Output Channel

Sensory Memory

Long-Term Memory (LTM)

Short-Term Memory (STM)

  • A. Cerone, Nazarbayev University – p.3/26
slide-5
SLIDE 5

Long-Term Memory (LTM)

CIFMA 2019, 17 September 2019

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

Short-Term Memory (STM) LTM Declarative

knowledge

  • f the world

“knowing what”

  • A. Cerone, Nazarbayev University – p.4/26
slide-6
SLIDE 6

Long-Term Memory (LTM)

CIFMA 2019, 17 September 2019

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

Short-Term Memory (STM) LTM Declarative

knowledge

  • f the world

“knowing what”

Procedural

refers to human skills “knowing how”

  • A. Cerone, Nazarbayev University – p.4/26
slide-7
SLIDE 7

LTM Components

CIFMA 2019, 17 September 2019

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

Short-Term Memory (STM) LTM Declarative

knowledge

  • f the world

“knowing what”

Procedural

refers to human skills “knowing how”

Episodic

events

  • A. Cerone, Nazarbayev University – p.5/26
slide-8
SLIDE 8

LTM Components

CIFMA 2019, 17 September 2019

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

Short-Term Memory (STM) LTM Declarative

knowledge

  • f the world

“knowing what”

Procedural

refers to human skills “knowing how”

Episodic

events

Semantic

facts

  • A. Cerone, Nazarbayev University – p.5/26
slide-9
SLIDE 9

Semantic Memory

CIFMA 2019, 17 September 2019

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

Short-Term Memory (STM) LTM Procedural

human skills

Episodic

events

Semantic

facts

  • A. Cerone, Nazarbayev University – p.6/26
slide-10
SLIDE 10

Semantic Memory

CIFMA 2019, 17 September 2019

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

Short-Term Memory (STM) LTM Procedural

human skills

Episodic

events

Semantic

facts semantic network

  • f associations

deliberate control processing

  • A. Cerone, Nazarbayev University – p.6/26
slide-11
SLIDE 11

Semantic Memory

CIFMA 2019, 17 September 2019

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

Short-Term Memory (STM) LTM Procedural

human skills

Episodic

events

Semantic

facts semantic network

  • f associations

deliberate control processing

  • A. Cerone, Nazarbayev University – p.6/26
slide-12
SLIDE 12

Example of Semantic Network

ANIMAL

does

breath

does

move

❍ ❍ ❍ ❍ ❨

is_a

DOG

does

bark

✟✟✟✟ ✯

has

four_legs

❍❍❍❍ ❥

has

tail

  • A. Cerone, Nazarbayev University – p.7/26
slide-13
SLIDE 13

Example of Semantic Network

ANIMAL

does

breath

does

move

❍ ❍ ❍ ❍ ❨

is_a

DOG

does

bark

✟✟✟✟ ✯

has

four_legs

❍❍❍❍ ❥

has

tail

is_a

SHEEPDOG

works

SHEEP

is_a

COLLIE

✻ is_a

LASSIE

is_a

HOUND

❍❍❍❍ ❥

does

track

is_a

BEAGLE

is_a

SNOOPY

is_a

BASENJI

✛doesnt

bark

  • A. Cerone, Nazarbayev University – p.7/26
slide-14
SLIDE 14

Example of Knowledge Domain

ANIMAL

does

breath

does

move

❍ ❍ ❍ ❍ ❨

is_a

DOG

does

bark

✟✟✟✟ ✯

has

four_legs

❍❍❍❍ ❥

has

tail

is_a

SHEEPDOG

works

SHEEP

is_a

COLLIE

is_a

LASSIE

is_a

HOUND

❍❍❍❍ ❥

does

track

is_a

BEAGLE

is_a

SNOOPY

is_a

BASENJI

✛doesnt

bark dogs

(knowledge domain)

  • A. Cerone, Nazarbayev University – p.8/26
slide-15
SLIDE 15

Example of Knowledge Domain

ANIMAL

does

breath

does

move

❍ ❍ ❍ ❍ ❨

is_a

DOG

does

bark

✟✟✟✟ ✯

has

four_legs

❍❍❍❍ ❥

has

tail

is_a

SHEEPDOG

works

SHEEP

is_a

COLLIE

is_a

LASSIE

is_a

HOUND

❍❍❍❍ ❥

does

track

is_a

BEAGLE

is_a

SNOOPY

is_a

BASENJI

✛doesnt

bark dogs

(knowledge domain)

animals

  • A. Cerone, Nazarbayev University – p.8/26
slide-16
SLIDE 16

Example of Knowledge Domain

ANIMAL

does

breath

does

move

❍ ❍ ❍ ❍ ❨

is_a

DOG

does

bark

✟✟✟✟ ✯

has

four_legs

❍❍❍❍ ❥

has

tail

is_a

SHEEPDOG

works

SHEEP

is_a

COLLIE

is_a

LASSIE

is_a

HOUND

❍❍❍❍ ❥

does

track

is_a

BEAGLE

is_a

SNOOPY

is_a

BASENJI

✛doesnt

bark dogs

(knowledge domain)

animals

character

cartoons

character

films

  • A. Cerone, Nazarbayev University – p.8/26
slide-17
SLIDE 17

Knowledge Representation

CIFMA 2019, 17 September 2019

animals : animal |

d1

= ⇒ | does(breath) ANIMAL

does

breath

does

move

  • A. Cerone, Nazarbayev University – p.9/26
slide-18
SLIDE 18

Knowledge Representation

CIFMA 2019, 17 September 2019

animals : animal |

d1

= ⇒ | does(breath) dogs : dog |

d2

= ⇒ | is_a(animal) ANIMAL

does

breath

does

move

❍ ❍ ❍ ❍ ❨

is_a

DOG

does

bark

✟✟✟✟ ✯

has

four_legs

❍❍❍❍ ❥

has

tail

  • A. Cerone, Nazarbayev University – p.9/26
slide-19
SLIDE 19

Knowledge Representation

CIFMA 2019, 17 September 2019

animals : animal |

d1

= ⇒ | does(breath) dogs : dog |

d2

= ⇒ | is_a(animal) dogs : dog |

d3

= ⇒ | does(bark) dogs : basenji |

d4

= ⇒ | doesnt(bark)

  • A. Cerone, Nazarbayev University – p.9/26
slide-20
SLIDE 20

Knowledge Representation

CIFMA 2019, 17 September 2019

animals : animal |

d1

= ⇒ | does(breath) dogs : dog |

d2

= ⇒ | is_a(animal) dogs : dog |

d3

= ⇒ | does(bark) dogs : basenji |

d4

= ⇒ | doesnt(bark) dogs : sheepdog |

d5

= ⇒ | is_a(dog) dogs : sheepdog |

d6

= ⇒ | works(sheep) dogs : collie |

d7

= ⇒ | is_a(sheepdog) dogs : lassie |

d8

= ⇒ | is_a(collie)

  • A. Cerone, Nazarbayev University – p.9/26
slide-21
SLIDE 21

Knowledge Representation

CIFMA 2019, 17 September 2019

animals : animal |

d1

= ⇒ | does(breath) dogs : dog |

d2

= ⇒ | is_a(animal) dogs : dog |

d3

= ⇒ | does(bark) dogs : basenji |

d4

= ⇒ | doesnt(bark) dogs : sheepdog |

d5

= ⇒ | is_a(dog) dogs : sheepdog |

d6

= ⇒ | works(sheep) dogs : collie |

d7

= ⇒ | is_a(sheepdog) dogs : lassie |

d8

= ⇒ | is_a(collie) dogs : hound |

d9

= ⇒ | does(track)

  • A. Cerone, Nazarbayev University – p.9/26
slide-22
SLIDE 22

Semantic Network in BRDL

CIFMA 2019, 17 September 2019

domain : category |

d

= ⇒ | type(attribute)

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

Short-Term Memory (STM) LTM Procedural

human skills

Episodic

events

Semantic

facts Example: dogs : dog |

d

= ⇒ | is_a(animal) semantic network

  • f associations

deliberate control processing

  • A. Cerone, Nazarbayev University – p.10/26
slide-23
SLIDE 23

Semantic Network in BRDL

CIFMA 2019, 17 September 2019

domain : category |

d

= ⇒ | type(attribute)

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

Short-Term Memory (STM) LTM Procedural

human skills

Episodic

events

Semantic

facts Example: dogs : dog |

d

= ⇒ | is_a(animal) dogs : dog |

d

= ⇒ | is_a(animal)

  • A. Cerone, Nazarbayev University – p.10/26
slide-24
SLIDE 24

Knowledge Retrieval

CIFMA 2019, 17 September 2019

domain : category |

d

= ⇒ | type(attribute) goal(domain, type?(category, attribute))

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

LTM

dogs : dog |

d

= ⇒ | is_a(animal)

Procedural

human skills

Episodic

events

Semantic

facts

Short-Term Memory (STM)

  • A. Cerone, Nazarbayev University – p.11/26
slide-25
SLIDE 25

Knowledge Retrieval

CIFMA 2019, 17 September 2019

domain : category |

d

= ⇒ | type(attribute) goal(domain, type?(category, attribute))

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

LTM

dogs : dog |

d

= ⇒ | is_a(animal)

Procedural

human skills

Episodic

events

Semantic

facts

STM

goal(dogs, is_a?(dog, animal)

  • A. Cerone, Nazarbayev University – p.11/26
slide-26
SLIDE 26

Knowledge Retrieval

CIFMA 2019, 17 September 2019

domain : category |

d

= ⇒ | type(attribute) goal(domain, type?(category, attribute))

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

LTM

dogs : dog |

d

= ⇒ | is_a(animal)

Procedural

human skills

Episodic

events

Semantic

facts

STM

goal(dogs, is_a?(dog, animal)

pattern matching

  • A. Cerone, Nazarbayev University – p.11/26
slide-27
SLIDE 27

Retrieval Outcome

CIFMA 2019, 17 September 2019

domain : category |

d

= ⇒ | type(attribute) goal(domain, type?(category, attribute))

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

LTM

dogs : dog |

d

= ⇒ | is_a(animal)

Procedural

human skills

Episodic

events

Semantic

facts

STM

is_a(dog, animal)

after time d (mental processing delay)

  • A. Cerone, Nazarbayev University – p.12/26
slide-28
SLIDE 28

Deeper Retrieval

CIFMA 2019, 17 September 2019

domain : category |

d

= ⇒ | type(attribute) goal(domain, type?(category, attribute))

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

LTM

dogs : dog |

d1

= ⇒ | is_a(animal) animal : animal |

d2

= ⇒ | does(breath)

Procedural

human skills

Episodic

events

Semantic

facts

STM

goal(dogs, does_a?(dog, breath)

  • A. Cerone, Nazarbayev University – p.13/26
slide-29
SLIDE 29

Deeper Retrieval

CIFMA 2019, 17 September 2019

domain : category |

d

= ⇒ | type(attribute) goal(domain, type?(category, attribute))

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

LTM

dogs : dog |

d1

= ⇒ | is_a(animal) animal : animal |

d2

= ⇒ | does(breath)

Procedural

human skills

Episodic

events

Semantic

facts

STM

goal(dogs, does_a?(dog, breath)

pattern matching

  • nly on breath

  • A. Cerone, Nazarbayev University – p.13/26
slide-30
SLIDE 30

Deeper Retrieval

CIFMA 2019, 17 September 2019

domain : category |

d

= ⇒ | type(attribute) goal(domain, type?(category, attribute))

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

LTM

dogs : dog |

d1

= ⇒ | is_a(animal) animal : animal |

d2

= ⇒ | does(breath)

Procedural

human skills

Episodic

events

Semantic

facts

STM

goal(dogs, does_a?(dog, breath)

pattern matching

  • nly on breath

✛ ✛

along the is_a hierarchy matching on animal

  • A. Cerone, Nazarbayev University – p.13/26
slide-31
SLIDE 31

Deeper Retrieval Outcome

CIFMA 2019, 17 September 2019

domain : category |

d

= ⇒ | type(attribute) goal(domain, type?(category, attribute))

Environment

Input Channel

Output Channel

Sensory Memory

✛ ❄

LTM

dogs : dog |

d1

= ⇒ | is_a(animal) animal : animal |

d2

= ⇒ | does(breath)

Procedural

human skills

Episodic

events

Semantic

facts

STM

does(dog, breath)

after time d1 + d2 (mental processing delay)

  • A. Cerone, Nazarbayev University – p.14/26
slide-32
SLIDE 32

Climbing the Semantic Network

ANIMAL

does

breath

does

move

❍ ❍ ❍ ❍ ❨

DOG

does

bark

✟✟✟✟ ✯

has

four_legs

❍❍❍❍ ❥

has

tail

is_a

SHEEPDOG

works

SHEEP

is_a

COLLIE

is_a

LASSIE

✻ ❍❍❍❍ ❥

does

track

is_a

BEAGLE

is_a

SNOOPY

✲ ✛doesnt

bark dogs

(knowledge domain)

animals

is_a

does(dog, breath) HOUND

is_a

BASENJI

is_a

  • A. Cerone, Nazarbayev University – p.15/26
slide-33
SLIDE 33

Climbing the Semantic Network

ANIMAL

does

breath

does

move

❍ ❍ ❍ ❍ ❨

DOG

does

bark

✟✟✟✟ ✯

has

four_legs

❍❍❍❍ ❥

has

tail

is_a

SHEEPDOG

works

SHEEP

is_a

COLLIE

is_a

LASSIE

✻ ❍❍❍❍ ❥

does

track

is_a

BEAGLE

is_a

SNOOPY

✲ ✛doesnt

bark dogs

(knowledge domain)

animals

is_a

does(basenji, breath) HOUND

is_a

BASENJI

is_a

  • A. Cerone, Nazarbayev University – p.15/26
slide-34
SLIDE 34

Deliberate Control: Reasoning

CIFMA 2019, 17 September 2019

in fer(domain) : prem |

d

= ⇒ | conseq

Environment

❄ ✻

Sensory Memory

✛ ❄

LTM

in fer(domain) : prem |

d

= ⇒ | conseq

Procedural

human skills

Episodic

events

Semantic

facts

STM

in fer(domain) prem . . .

  • A. Cerone, Nazarbayev University – p.16/26
slide-35
SLIDE 35

Deliberate Control: Reasoning

CIFMA 2019, 17 September 2019

in fer(domain) : prem |

d

= ⇒ | conseq

Environment

❄ ✻

Sensory Memory

✛ ❄

LTM

in fer(domain) : prem |

d

= ⇒ | conseq

Procedural

human skills

Episodic

events

Semantic

facts

STM

conseq prem . . .

  • A. Cerone, Nazarbayev University – p.16/26
slide-36
SLIDE 36

Reasoning Example

CIFMA 2019, 17 September 2019

in fer(dr) : {zebra, ped} |

d

= ⇒ | goal(dr, {gw}) Example: Give way to pedestrians who are ready to walk across a zebra crossing

Environment

❄ ✻

Sensory Memory

✛ ❄

LTM

in fer(dr) : {zebra, ped} |

d

= ⇒ | goal(dr, {gw)}

Procedural

human skills

Episodic

events

Semantic

facts

STM

in fer(dr) {zebra, ped} . . .

  • A. Cerone, Nazarbayev University – p.17/26
slide-37
SLIDE 37

Reasoning Example

CIFMA 2019, 17 September 2019

in fer(dr) : {zebra, ped} |

d

= ⇒ | goal(dr, {gw}) Example: Give way to pedestrians who are ready to walk across a zebra crossing

Environment

❄ ✻

Sensory Memory

✛ ❄

LTM

in fer(dr) : {zebra, ped} |

d

= ⇒ | goal(dr, {gw)}

Procedural

human skills

Episodic

events

Semantic

facts

STM

goal(dr, {gw}) zebra, ped . . .

  • A. Cerone, Nazarbayev University – p.17/26
slide-38
SLIDE 38

Reasoning Example

CIFMA 2019, 17 September 2019

in fer(dr) : {zebra, ped} |

d

= ⇒ | goal(dr, {gw}) Example: Give way to pedestrians who are ready to walk across a zebra crossing

Environment

❄ ✻

Sensory Memory

✛ ❄

LTM

in fer(dr) : {zebra, ped} |

d

= ⇒ | goal(dr, {gw)}

Procedural

human skills

Episodic

events

Semantic

facts

STM

goal(dr, {gw}) zebra, ped . . . zebra, ped

  • A. Cerone, Nazarbayev University – p.17/26
slide-39
SLIDE 39

Reasoning Example

CIFMA 2019, 17 September 2019

in fer(dr) : {zebra, ped} |

d

= ⇒ | goal(dr, {gw}) Example: Give way to pedestrians who are ready to walk across a zebra crossing

Environment

❄ ✻

Sensory Memory

✛ ❄

LTM

in fer(dr) : {zebra, ped} |

d

= ⇒ | goal(dr, {gw)}

Procedural

human skills

Episodic

events

Semantic

facts

STM

goal(dr, {gw}) zebra, ped . . . zebra, ped implement the rule gw

  • A. Cerone, Nazarbayev University – p.17/26
slide-40
SLIDE 40

Deliberate Control: Behaviour

CIFMA 2019, 17 September 2019

goal : in fo1 ↑ perc

d

= ⇒ act ↓ in fo2

Environment

perc

Sensory Memory

✛ ❄

goal . . .

STM LTM Procedural

human skills

Episodic

events

Semantic

facts goal : in fo1 perc

d

= ⇒ act in fo2 selected in fo

✲ ✻

in fo1 . . . perc

  • A. Cerone, Nazarbayev University – p.18/26
slide-41
SLIDE 41

Deliberate Control: Behaviour

CIFMA 2019, 17 September 2019

goal : in fo1 ↑ perc

d

= ⇒ act ↓ in fo2

Environment

perc

Sensory Memory

✛ ❄

goal . . .

STM LTM Procedural

human skills

Episodic

events

Semantic

facts goal : in fo1 perc

d

= ⇒ act in fo2 selected in fo

✲ ✻

after delay d (mental processing) in fo2 . . . act

  • A. Cerone, Nazarbayev University – p.18/26
slide-42
SLIDE 42

Goal Achievement

CIFMA 2019, 17 September 2019

Environment

perc

act

Sensory Memory

✛ ❄

goal(dom, in fo) . . .

STM LTM Procedural

human skills

Episodic

events

Semantic

facts goal(dom, in fo) : in fo1 ↓ perc

d

= ⇒ act ↑ in fo2 selected in fo perc

✲ ✻

goal(dom, in fo) : in fo1 ↑ perc

d

= ⇒ act ↓ in fo2 in fo = ∅ and in fo⊆{perc,act}∪ in fo2∪ in foR in fo1 ∪ in foR

  • A. Cerone, Nazarbayev University – p.19/26
slide-43
SLIDE 43

Goal Achievement

CIFMA 2019, 17 September 2019

Environment

perc

act

Sensory Memory

✛ ❄

goal(dom, in fo) . . .

STM LTM Procedural

human skills

Episodic

events

Semantic

facts goal(dom, in fo) : in fo1 ↓ perc

d

= ⇒ act ↑ in fo2 selected in fo perc

✲ ✻

goal(dom, in fo) : in fo1 ↑ perc

d

= ⇒ act ↓ in fo2 in fo = ∅ and in fo⊆{perc,act}∪ in fo2∪ in foR in fo1 ∪ in foR

  • A. Cerone, Nazarbayev University – p.19/26
slide-44
SLIDE 44

Closure

CIFMA 2019, 17 September 2019

Environment

perc

act

Sensory Memory

✛ ❄

goal(dom, in fo) . . .

STM LTM Procedural

human skills

Episodic

events

Semantic

facts goal(dom, in fo) : in fo1 ↓ perc

d

= ⇒ act ↑ in fo2 selected in fo perc

✲ ✻

goal(dom, in fo) : in fo1 ↑ perc

d

= ⇒ act ↓ in fo2 in fo = ∅ and in fo⊆{perc,act}∪ in fo2∪ in foR in fo2 ∪ cloMod(in foR)

✲ Closure ❄ ✟ ✟ ✯

  • A. Cerone, Nazarbayev University – p.20/26
slide-45
SLIDE 45

= ⇒ Automatic Control

CIFMA 2019, 17 September 2019

goal(dom, in fo) : in fo1 ↑ perc

d

= ⇒ act ↓ in fo2 in fo1 ↑ perc

d

= ⇒ act ↓ in fo2

Environment

perc

act

Sensory Memory

✛ ❄

goal(in fo)

STM LTM Procedural

human skills

Episodic

events in fo1 ↓ perc

d

= ⇒ act ↑ in fo2

Semantic

facts goal(dom, in fo) : in fo1 ↓ perc

d

= ⇒ act ↑ in fo2 selected in fo in fo1 ∪ in foR perc

✲ ✻ ✲

Skill Acquisition Closure

❄ ✟✟ ✟ ✯ ✻ ✻

Explicit Attention Implicit Attention ✻

  • A. Cerone, Nazarbayev University – p.21/26
slide-46
SLIDE 46

Deliberate Behaviour Example

CIFMA 2019, 17 September 2019

goal(dr, {ped}) : zebra ↑ ped

d

= ⇒ ↓ zebra, ped, in fer(dr)

Environment

ped

Sensory Memory

✛ ❄

STM LTM Procedural

human skills

Episodic

events

Semantic

facts goal(dr, {ped}) : zebra | ped

d

= ⇒ ↓ ped, in fer(dr) selected in fo

ped goal(dr, {ped}), zebra

  • A. Cerone, Nazarbayev University – p.22/26
slide-47
SLIDE 47

Deliberate Behaviour Example

CIFMA 2019, 17 September 2019

goal(dr, {ped}) : zebra ↑ ped

d

= ⇒ ↓ zebra, ped, in fer(dr)

Environment

ped

Sensory Memory

✛ ❄

STM LTM Procedural

human skills

Episodic

events

Semantic

facts goal(dr, {ped}) : zebra | ped

d

= ⇒ ↓ ped, in fer(dr) selected in fo

✲ ✟✟✟ ✟ ✯

zebra in fer(dr), ped

  • A. Cerone, Nazarbayev University – p.22/26
slide-48
SLIDE 48

Behaviour+Reasoning

CIFMA 2019, 17 September 2019

in fer(dr) : zebra, ped |

d

= ⇒ | goal(dr, {gw}) Example: Give way to pedestrians who are ready to walk across a zebra crossing

Environment

❄ ✻

Sensory Memory

✛ ❄

LTM

in fer(dr) : zebra, ped |

d

= ⇒ | goal(dr, {gw})

Procedural

human skills

Episodic

events

Semantic

facts

STM

zebra in fer(dr), ped

❅ ❅ ❅ ❅ ❅ ■

  • A. Cerone, Nazarbayev University – p.23/26
slide-49
SLIDE 49

Behaviour+Reasoning

CIFMA 2019, 17 September 2019

in fer(dr) : zebra, ped |

d

= ⇒ | goal(dr, {gw}) Example: Give way to pedestrians who are ready to walk across a zebra crossing

Environment

❄ ✻

Sensory Memory

✛ ❄

LTM

in fer(dr) : zebra, ped |

d

= ⇒ | goal(dr, {gw})

Procedural

human skills

Episodic

events

Semantic

facts

STM

zebra goal(dr, {gw}), ped

✟✟✟ ✟ ✯

  • A. Cerone, Nazarbayev University – p.23/26
slide-50
SLIDE 50

Behaviour+Reasoning: Action

CIFMA 2019, 17 September 2019

goal(dr, {gw}) : ∅ ↑

d

= ⇒ stop ↓ gw Example: Give way to pedestrians who are ready to walk across a zebra crossing

Environment

❄ ✻

Sensory Memory

✛ ❄

LTM

goal(dr, {gw}) : ∅ ↓

d

= ⇒ stop ↓ gw

Procedural

human skills

Episodic

events

Semantic

facts

STM

zebra goal(dr, gw), ped

❅ ■

  • A. Cerone, Nazarbayev University – p.24/26
slide-51
SLIDE 51

Behaviour+Reasoning: Action

CIFMA 2019, 17 September 2019

goal(dr, {gw}) : ∅ ↑

d

= ⇒ stop ↓ gw Example: Give way to pedestrians who are ready to walk across a zebra crossing

Environment

❄ ✻

Sensory Memory

✛ ❄

LTM

goal(dr, {gw}) : ∅ ↓

d

= ⇒ stop ↓ gw

Procedural

human skills

Episodic

events

Semantic

facts

STM

gw, ped

✟ ✟ ✯ ✻ ❄

Closure

  • A. Cerone, Nazarbayev University – p.24/26
slide-52
SLIDE 52

Conclusion

CIFMA 2019, 17 September 2019

Behaviour and Reasoning Description Language

  • knolwledge representation
  • knowledge retrieval
  • reasoning
  • human behaviour
  • problem solving
  • A. Cerone, Nazarbayev University – p.25/26
slide-53
SLIDE 53

Conclusion

CIFMA 2019, 17 September 2019

Behaviour and Reasoning Description Language

  • knolwledge representation
  • knowledge retrieval
  • reasoning
  • human behaviour
  • problem solving
  • A. Cerone, Nazarbayev University – p.25/26
slide-54
SLIDE 54

Further and Future Work

CIFMA 2019, 17 September 2019

Further Completed Work

  • Implementation of reasoning

Antonio Cerone and Peter Csaba Ölveczky Modelling Human Reasoning in Practical Behavioural Contexts using Real-time Maude to be presented at FMIS 2019, 7 October 2019, Porto, Portugal

  • A. Cerone, Nazarbayev University – p.26/26
slide-55
SLIDE 55

Further and Future Work

CIFMA 2019, 17 September 2019

Further Completed Work

  • Implementation of reasoning

Antonio Cerone and Peter Csaba Ölveczky Modelling Human Reasoning in Practical Behavioural Contexts using Real-time Maude to be presented at FMIS 2019, 7 October 2019, Porto, Portugal Future Work

  • Implementation of problem solving
  • A. Cerone, Nazarbayev University – p.26/26
slide-56
SLIDE 56

Further and Future Work

CIFMA 2019, 17 September 2019

Further Completed Work

  • Implementation of reasoning

Antonio Cerone and Peter Csaba Ölveczky Modelling Human Reasoning in Practical Behavioural Contexts using Real-time Maude to be presented at FMIS 2019, 7 October 2019, Porto, Portugal Future Work

  • Implementation of problem solving
  • Use the language to
  • compare alternative theories of cognition
  • formally verify interactive systems
  • A. Cerone, Nazarbayev University – p.26/26