HUMAN-COMPUTER CO-CREATION Anna Kantosalo - - PowerPoint PPT Presentation

▶
human computer co creation
SMART_READER_LITE
LIVE PREVIEW

HUMAN-COMPUTER CO-CREATION Anna Kantosalo - - PowerPoint PPT Presentation

Dec 04, 2022 381 likes •727 views

HUMAN-COMPUTER CO-CREATION Anna Kantosalo Matemaattis-luonnontieteellinen tiedekunta CC-2017 Anna Kantosalo 24/11/2017 1 OUTLINE DEFINITION AIMS AND SCOPE ROLES MODELING HUMAN COMPUTER CO-CREATION DESIGNING HUMAN COMPUTER CO-CREATION

slide-1
SLIDE 1

Matemaattis-luonnontieteellinen tiedekunta

Anna Kantosalo

HUMAN-COMPUTER CO-CREATION

24/11/2017 CC-2017 Anna Kantosalo 1

slide-2
SLIDE 2

Matemaattis-luonnontieteellinen tiedekunta

OUTLINE

DEFINITION AIMS AND SCOPE ROLES MODELING HUMAN COMPUTER CO-CREATION DESIGNING HUMAN COMPUTER CO-CREATION

24/11/2017 CC-2017 Anna Kantosalo 2

slide-3
SLIDE 3

Matemaattis-luonnontieteellinen tiedekunta

DEFINITION

Human-computer co-creativity is collaborative creativity where both the human and the computer take creative responsibility for the generation of a creative artefact.

  • Kantosalo, Toivanen, Xiao and Toivonen, 2014

24/11/2017 CC-2017 Anna Kantosalo 3

slide-4
SLIDE 4

Matemaattis-luonnontieteellinen tiedekunta

DEFINITION

Human-computer co-creativity is collaborative creativity where both the human and the computer take creative responsibility for the generation of a creative artefact.

  • Kantosalo, Toivanen, Xiao and Toivonen, 2014

24/11/2017 CC-2017 Anna Kantosalo 4

slide-5
SLIDE 5

Matemaattis-luonnontieteellinen tiedekunta

DEFINITION

Another frequently used definition is for Mixed-Initiative Co- Creativity:

” the task of creating artifacts via the interaction of a human initiative and a computational initiative.”

– Yannakakis, Liapis and Alexopoulos, 2014

24/11/2017 CC-2017 Anna Kantosalo 5

slide-6
SLIDE 6

Matemaattis-luonnontieteellinen tiedekunta

DEFINITION

Another frequently used definition is for Mixed-Initiative Co- Creativity:

” the task of creating artifacts via the interaction of a human initiative and a computational initiative.”

– Yannakakis, Liapis and Alexopoulos, 2014

24/11/2017 CC-2017 Anna Kantosalo 6

slide-7
SLIDE 7

Matemaattis-luonnontieteellinen tiedekunta

DEFINITION

24/11/2017 CC-2017 Anna Kantosalo 7

10 10

slide-8
SLIDE 8

Matemaattis-luonnontieteellinen tiedekunta

DEFINITION

24/11/2017 Presentation Name / Firstname Lastname 8

10 10 10 10 10 10

slide-9
SLIDE 9

Matemaattis-luonnontieteellinen tiedekunta

AIMS

The goal of human-computer co-creativity is to enhance both, human and computational creativity.

24/11/2017 CC-2017 Anna Kantosalo 9

slide-10
SLIDE 10

Matemaattis-luonnontieteellinen tiedekunta

AIMS

The goal of human-computer co-creativity is to enhance both, human and computational creativity.

  • Support for both, human and computational creativity
  • E.g. distributing tasks by strengths: Human does what humans are

good at, and computer does what computers are good at

24/11/2017 CC-2017 Anna Kantosalo 10

slide-11
SLIDE 11

Matemaattis-luonnontieteellinen tiedekunta

AIMS

The goal of human-computer co-creativity is to enhance both, human and computational creativity.

  • Support for both, human and computational creativity
  • E.g. distributing tasks by strengths: Human does what humans are

good at, and computer does what computers are good at

  • Creating new ways to create
  • E.g. new intelligent digital instruments for musicians

24/11/2017 CC-2017 Anna Kantosalo 11

slide-12
SLIDE 12

Matemaattis-luonnontieteellinen tiedekunta

SCOPE

24/11/2017 CC-2017 Anna Kantosalo 12

Computational Creativity

Human- Computer Co- Creativity

Creativity Support Systems Psychology

  • f Creativity

Human- Computer Interaction

Interaction Design Social Creativity

slide-13
SLIDE 13

Matemaattis-luonnontieteellinen tiedekunta

SCOPE

Application domains vary from recreational to serious applications: Serious examples:

  • Fields: Design, Interior Design, Architecture, Education…
  • DarwinsGaze (DiPaola and Gabora, 2009), Poetry Machine

(Kantosalo, Toivanen, Xiao and Toivonen, 2014) Recreational examples:

  • Games, Maker-culture…
  • Drawing Apprentice (Davis, Hsiao, Singh, Li and Magerko, 2016)

24/11/2017 CC-2017 Anna Kantosalo 13

slide-14
SLIDE 14

Matemaattis-luonnontieteellinen tiedekunta

ROLES

How do we start about discussing human-computer co-creation?

24/11/2017 CC-2017 Anna Kantosalo 14

10 10

slide-15
SLIDE 15

Matemaattis-luonnontieteellinen tiedekunta

ROLES

24/11/2017 CC-2017 Anna Kantosalo 15

Lubart (2005): How can computers be partners in the creative process

  • Nanny
  • Support work activity
  • Pen-Pal
  • Feedback manager
  • Coach
  • Technique trainer
  • Colleague
  • Autonomous creator
slide-16
SLIDE 16

Matemaattis-luonnontieteellinen tiedekunta

ROLES

24/11/2017 CC-2017 Anna Kantosalo 16

Lubart (2005): How can computers be partners in the creative process

  • Nanny
  • Support work activity
  • Pen-Pal
  • Feedback manager
  • Coach
  • Technique trainer
  • Colleague
  • Autonomous creator

Maher (2012): Who’s being creative?

  • Support
  • Tools and techniques
  • Enhance
  • Extend abilities
  • Generate
  • Generate artefacts

+ Human roles: Model and Generate

slide-17
SLIDE 17

Matemaattis-luonnontieteellinen tiedekunta

ROLES

24/11/2017 CC-2017 Anna Kantosalo 17

Lubart (2005): How can computers be partners in the creative process

  • Nanny
  • Support work activity
  • Pen-Pal
  • Feedback manager
  • Coach
  • Technique trainer
  • Colleague
  • Autonomous creator

Maher (2012): Who’s being creative?

  • Support
  • Tools and techniques
  • Enhance
  • Extend abilities
  • Generate
  • Generate artefacts

+ Human roles: Model and Generate Nakakoji (2006): Meanings

  • f tools, support, and uses of

creative design processes

  • Dumbell
  • Training creativity
  • Running Shoes
  • Creating faster
  • Skis
  • Creating something

different

slide-18
SLIDE 18

Matemaattis-luonnontieteellinen tiedekunta

MODELLING HCCC

ITERATIVE CO-CREATIVITY AS A SEARCH

  • To be able to discuss problems and difficulties in human

computer co-creation from a computational perspective a computational model is needed.

  • The Alternating co-creativity model extends Wiggins’ Creative

Systems Framework to an iterative co-creative scenario

  • Look for full definition in article Kantosalo and Toivonen 2016

24/11/2017 CC-2017 Anna Kantosalo 18

slide-19
SLIDE 19

Matemaattis-luonnontieteellinen tiedekunta

MODELLING HCCC

ITERATIVE CO-CREATIVITY AS A SEARCH

  • It is described for two participants, but in principle, it scales up

indefinitely

  • When reduced to the case of one praticipant, it becomes

Wiggins’ original model

  • It assumes indirect interaction via the generated artefact!

24/11/2017 CC-2017 Anna Kantosalo 19

10 10

slide-20
SLIDE 20

Matemaattis-luonnontieteellinen tiedekunta

MODELLING HCCC

ITERATIVE CO-CREATIVITY AS A SEARCH

U U

Sets of all possible concepts that the human and the computer can process

R R

Sets of valid concepts

E E

Sets of appreciated concepts

T

(c) T c

Sets of concepts reachable in n steps from c

t (c) t

The concept produced after c The above sets are defined by the respective rules ,,,,,

  • 24/11/2017

CC-2017 Anna Kantosalo 20

slide-21
SLIDE 21

Matemaattis-luonnontieteellinen tiedekunta

MODELLING HCCC

ITERATIVE CO-CREATIVITY AS A SEARCH

c

= t c

  • c

= t c

  • 24/11/2017

CC-2017 Anna Kantosalo 21

The co-creative process is described by cycles of two

  • perations:

The computational participant computes the ith iteration of the artefact from the (i- 1)th iteration of the artefact, produced by the human participant. The human participant computes the (i+1)th iteration of the artefact from the ith iteration of the artefact, produced by the human participant.

slide-22
SLIDE 22

Matemaattis-luonnontieteellinen tiedekunta

MODELLING HCCC

ITERATIVE CO-CREATIVITY AS A SEARCH

c

= t c

  • c

= t c

  • 24/11/2017

CC-2017 Anna Kantosalo 22

The co-creative process is described by cycles of two

  • perations:

The computational participant computes the ith iteration of the artefact from the (i- 1)th iteration of the artefact, produced by the human participant. The human participant computes the (i+1)th iteration of the artefact from the ith iteration of the artefact, produced by the human participant.

But wait! Creativity is supposed to produce both valid, and valuable

  • bjects!
slide-23
SLIDE 23

Matemaattis-luonnontieteellinen tiedekunta

MODELLING HCCC

ITERATIVE CO-CREATIVITY AS A SEARCH

24/11/2017 CC-2017 Anna Kantosalo 23

Valid of artefacts in Wiggins’ framework are described with set R. Valid artefacts in co-creation: R ∩ R Appreciated artefacts in Wiggins’ framework are described with E. Valid artefacts in co-creation: E ∩ E

slide-24
SLIDE 24

Matemaattis-luonnontieteellinen tiedekunta

MODELLING HCCC

ITERATIVE CO-CREATIVITY AS A SEARCH

24/11/2017 CC-2017 Anna Kantosalo 24

Valid of artefacts in Wiggins’ framework are described with set R. Valid artefacts in co-creation: R ∩ R Appreciated artefacts in Wiggins’ framework are described with E. Valid artefacts in co-creation: E ∩ E What if the computer doesn’t appreciate any concepts considered valid by the human? E.g. E ∩ R = ∅

slide-25
SLIDE 25

Matemaattis-luonnontieteellinen tiedekunta

MODELLING HCCC

ITERATIVE CO-CREATIVITY AS A SEARCH

24/11/2017 CC-2017 Anna Kantosalo 25

Universal Mismatch c

¬∈ U or

c

¬∈ U

Conceptual Mismatch

c

¬∈ R or

c

¬∈ R

Artistic Disagreement

c

¬∈ E or

c

¬∈ E

Generative Impotence

T

c

  • = ∅ or

T

c

  • = ∅
slide-26
SLIDE 26

Matemaattis-luonnontieteellinen tiedekunta

MODELLING HCCC

ITERATIVE CO-CREATIVITY AS A SEARCH

24/11/2017 CC-2017 Anna Kantosalo 26

How to tackle the problem?

  • Allow ”skipping turns” → Asymmetric alternating co-creativity
  • Transformational creativity → Change ,,

(or ,,) to

allow the understanding and evaluation of an artefact

  • In the case of Artistic Disagreement, the computer can also give

up on its own creative goals.

  • Finally, we can separate the creative tasks under Wiggins’ model

and participate only partially, via defining the conceptual space,

  • r value of the artefacts, or by only generating them. This is

called Task-divided co-creativity

slide-27
SLIDE 27

Matemaattis-luonnontieteellinen tiedekunta

MODELLING HCCC

ITERATIVE CO-CREATIVITY AS A SEARCH

24/11/2017 CC-2017 Anna Kantosalo 27

Ultimately, the strategy chosen to deal with Universal Mismatch, Conceptual Mismatch, Artistic Disagreement and Generative Impotence affects who controls the creation and whose artistic vision is pursued in alternating co-creativity. From computational perspective the role of the computational participant can therefore be categorised as complete vs. incomplete and pleasing vs. provoking, depending on which strategies are chosen to solve the problems!

slide-28
SLIDE 28

Matemaattis-luonnontieteellinen tiedekunta

MODELLING HCCC

ITERATIVE CO-CREATIVITY AS A SEARCH

24/11/2017 CC-2017 Anna Kantosalo 28

Incomplete co-creative agents

  • Are only capable of some,

not all creative tasks

  • Can only participate in task-

divided co-creativity

  • Most current co-creative

systems, e.g. evolutionary systems where human acts as the evaluator Complete co-creative agents

  • Are capable of identifying,

generating, and evaluating some concepts in a space

  • Can participate in

symmetric alternating co- creativity if transformational

  • Currently best fit: Tanagra

(Smith, Whitehead, and Mateas 2010)

slide-29
SLIDE 29

Matemaattis-luonnontieteellinen tiedekunta

MODELLING HCCC

ITERATIVE CO-CREATIVITY AS A SEARCH

24/11/2017 CC-2017 Anna Kantosalo 29

Provoking co-creative agents:

  • Provoking agents resist

changes to their rulesets based on human preferences alone

  • May outright challenge the

human creator and in general may be difficult to work with Pleasing co-creative agents:

  • A pleasing agent seeks to

modify its behavior by transforming its rulesets to better comply with the human’s needs

  • A pleasing agent may

essentially limit its own creativity

slide-30
SLIDE 30

Matemaattis-luonnontieteellinen tiedekunta

MODELLING HCCC

ITERATIVE CO-CREATIVITY AS A SEARCH

24/11/2017 CC-2017 Anna Kantosalo 30

Lack of communication is a general challenge for this model. With another communication channel

  • Human and computer could better agree how to make

creative transformations

  • Search could be directed to specific alternative routes via

discussion

  • The co-creative partners could better agree when to stop
  • Evidence suggests that human co-creators also work

somewhat iteratively, but their discussions often go to the meta-level, for example discussing alternate artefacts prior to settling to a specific change made to the artifact

slide-31
SLIDE 31

Matemaattis-luonnontieteellinen tiedekunta

DESIGNING HCCC

In designing human-computer co-creative applications, the skills, needs, and goals of the human participant become an integral part

  • f the design process.

Interaction Design and Human-Computer Interaction can be utilized to achieve good results. Sometimes the goals of the human users and computational creativity creators can be in contrast!

  • Algorithms may need to be dumbed or tuned down to achieve better

results

  • Usually humans are given more responsibility in the creation: They
  • ften decide the search space and end the search when they are

happy with the results

24/11/2017 CC-2017 Anna Kantosalo 31

slide-32
SLIDE 32

Matemaattis-luonnontieteellinen tiedekunta

CONCLUSIONS

  • Human-computer co-creativity aims to enhance human and

computational creativity

  • The human and the computational co-creator can take different roles

within the co-creative process

  • These roles usually describe the purpose of the participant, but can

also explain the capability of the participant within co-creation

  • Computational models are important as they can help to recognize

important pitfalls when designing systems

  • So far a focus on interaction and communication between the co-

creative participants has received little focus

  • Designing HCCC has so far focused on responding to human needs

and identifying what role the system can take within the context

24/11/2017 CC-2017 Anna Kantosalo 32

slide-33
SLIDE 33

Matemaattis-luonnontieteellinen tiedekunta

REFERENCES

DiPaola, S.; McCaig, G.; Carlson, K.; Salevati, S.; and Sorenson, N. 2013. Adaptation of an autonomous creative evolutionary system for real-world design application based on creative cognition. In Proceedings of the Fourth International Conference on Computational Creativity. June 12-14, 2013, Sydney, Australia, 40–47. Kantosalo, A. and Toivonen, H. 2016. Modes for Creative Human-Computer Collaboration: Alternating and Task-Divided Co-

  • Creativity. In Proceedings of the 7th International Conference on Computational Creativity, June 27- July 1, 2016, Paris, 77-

84. Kantosalo, A.; Toivanen, J. M.; Xiao, P.; and Toivonen, H. 2014. From isolation to involvement: Adapting machine creativity software to support human-computer cocreation. In Proceedings of the Fifth International Conference on Computational

  • Creativity. June 10-13, 2014, Ljubljana, Slovenia, 1–8.

Lubart, T. 2005. How can computers be partners in the creative process: classification and commentary on the special issue. International Journal of Human-Computer Studies 63(4):365–369. Maher, M. L. 2012. Computational and collective creativity: whos being creative? In Proceedings of the Third International Conference on Computational Creativity. May 30 -June 1, 2012. Dublin, Ireland, 67–71. Nakakoji, K. 2006. Meanings of tools, support, and uses for creative design processes. In International Design Research Symposium ’06, Seoul, Korea, 156–165. Smith, G.; Whitehead, J.; and Mateas, M. 2010. Tanagra: A mixed-initiative level design tool. In Proceedings of the Fifth International Conference on the Foundations of Digital Games. June 19-21, 2010. Monterey, California, USA, FDG ’10, 209–

  • 216. New York, NY, USA: ACM.

Wiggins, G. A. 2006. A preliminary framework for description, analysis and comparison of creative systems. Knowledge- based Systems 19(7):449–458. Yannakakis, G. N.; Liapis, A.; and Alexopoulos, C. 2014. Mixed-initiative co-creativity. In Proceedings of the 9th Conference

  • n the Foundations of Digital Games. April 3-7. 2014. Ft. Lauderdale, Florida, USA.

24/11/2017 CC-2017 Anna Kantosalo 33