cec 2009: towards creative design using collaborative interactive genetic algorithms
DESCRIPTION
CEC 2009 presentation.TRANSCRIPT
Towards Creative Design Using Collaborative Interactive Genetic
Algorithms
Juan C. Quiroz, Sushil J. Louis, Amit Banerjee, and Sergiu Dascalu
Evolutionary Computing Systems LabDepartment of Computer Science & Engineering
University of Nevada, Reno
CEC 2009 [email protected] 2
Outline
• Motivation• Computational model of creativity• Design space exploration• IGAP• Experimental Setup• Discussion and Results
CEC 2009 [email protected] 3
Creativity
• John Gero’s definition of creative designing:– The addition of variables during the design
process has the potential, but does not guarantee, to generate creative content.
• Is collaboration amongst peers sufficient to allow for the potential to produce creative content in designing?
CEC 2009 [email protected] 4
Motivation
• Design process1. Conceptual design2. Detailed design3. Evaluation4. Iterative redesign
CEC 2009 [email protected] 5
Conceptual Design
• Subjective evaluation of alternative design concepts– Aesthetics and other subjective criteria
• What is the formula for how designers evaluate subjective criteria?
• Collaborative design
CEC 2009 [email protected] 7
Computational Model of Creative Design:Collaborative Interactive Genetic Algorithms
CEC 2009 [email protected] 8
Design Space Exploration
CEC 2009 [email protected] 9
IGAP: Interactive GA Peer to Peer
CEC 2009 [email protected] 12
Experimental Setup
• Objectives:– Create a floorplan for a 2 bedroom, 1 bathroom
apartment– Bathrooms close to the bedrooms– Bathrooms far from kitchen and dining areas
CEC 2009 [email protected] 13
Experimental Setup
• Participants:– 8 women, 12 men
• Five groups of size four• Agenda
1. Tutorial2. Create individual floorplan3. Create collaborative floorplan4. Evaluation of floorplans
CEC 2009 [email protected] 14
Evaluation Criteria1. Appealing – unappealing2. Average – revolutionary3. Commonplace – original4. Conventional – unconventional5. Dull – exciting6. Fresh - routine7. Novel – predictable8. Unique – ordinary9. Usual - unusual10. Meets all requirements - does
not meet requirements
• Creative Product Semantic Scale• Seven point likert scale
CEC 2009 [email protected] 15
Hypothesis
• Is collaboration amongst peers sufficient to allow for the potential to produce creative content in designing?
• Designs evolved collaboratively will consistently rank higher in the evaluation criteria.
CEC 2009 [email protected] 16
ResultsEvaluation Criterion Desired Ind. Avg. Coll. Avg. P-value
Appealing - Unappealing Low 4.08 4.39 0.439
Average - Revolutionary High 3.76 4.34 0.047Commonplace - Original High 3.97 4.68 0.021
Conventional - Unconventional
High 4.03 4.41 0.355
Dull – Exciting High 3.65 3.93 0.326
Fresh – Routine Low 3.82 3.68 0.810
Novel - Predictable Low 3.55 3.40 0.697Unique - Ordinary Low 3.49 3.11 0.251Usual - Unusual High 4.21 4.51 0.395
Meets All Req. -Does Not Meet Req.
Low 2.63 2.83 0.779
CEC 2009 [email protected] 17
Preliminary Conclusions
• Collaboration is not sufficient to make a clear distinction between individual and collaborative floorplans (?)
CEC 2009 [email protected] 18
Observations
• Ambiguity in evaluation criteria– Appealing – unappealing– Positive – Negative (?)– Negative – Positive (?)
• Simple graphic representation
CEC 2009 [email protected] 19
Observations: A New Hope
• Applicability of evaluation criteria– “Exciting”– Domain expert vs. student
• Participants created only 1 collaborative floorplan and 1 individual floorplan
• Negative effect of viewing entire population– Not enough generations– Not enough sharing
CEC 2009 [email protected] 20
Conclusions
• Collaborative IGAs as a computational model of creative design
• Floorplans created collaboratively scored higher in “originality” and “revolutionary”– Scores in median range– Graphical representation
• We cannot deduce that collaboration is enough to introduce a creative potential
CEC 2009 [email protected] 21
Comments/Questions?
• [email protected]• www.cse.unr.edu/~quiroz
• Acknowledgements– This work was supported in part by contract number
N00014-0301-0104 from the Office of Naval Research and the National Science Foundation under Grant no. 0447416.