65
Creative Design Using Collaborative Interactive Genetic Algorithms Juan C. Quiroz PhD Dissertation Defense Thursday April 29, 2010 Department of Computer Science & Engineering University of Nevada, Reno

Creative Design Using Collaborative Interactive Genetic Algorithms

Embed Size (px)

DESCRIPTION

Dissertation defense. I propose a computational model of creative design based on collaborative interactive genetic algorithms. I test the computational model on two case studies: floorplanning and 3D modeling.

Citation preview

Page 1: Creative Design Using Collaborative Interactive Genetic Algorithms

Creative Design Using Collaborative Interactive Genetic Algorithms

Juan C. QuirozPhD Dissertation DefenseThursday April 29, 2010

Department of Computer Science & EngineeringUniversity of Nevada, Reno

Page 2: Creative Design Using Collaborative Interactive Genetic Algorithms

Outline

1. Creativity in Design

2. Collaborative Interactive Genetic Algorithms

3. Reducing User Fatigue in Interactive Genetic Algorithms

4. Testing Our Computational Model of Creative Design

5. Contributions

Page 3: Creative Design Using Collaborative Interactive Genetic Algorithms

Design

• Conceptual design• Detailed design• Evaluation • Iterative redesign

Page 4: Creative Design Using Collaborative Interactive Genetic Algorithms

Conceptual Design

• Initially conceiving and elaborating solutions that meet a set of requirements

• Change in requirements• Subjective evaluation of alternative design

concepts– Aesthetics and other subjective criteria

• Collaboration

Page 5: Creative Design Using Collaborative Interactive Genetic Algorithms

Creativity

• Novel and useful• Role of collaboration

Individual Work

Individual

PeersWork

Page 6: Creative Design Using Collaborative Interactive Genetic Algorithms

Computational Model of Creative Design

• Allows for subjective exploration of solutions• Supports collaboration• Has the potential to generate creative

solutions

Page 7: Creative Design Using Collaborative Interactive Genetic Algorithms

Main Claim

Collaborative interactive genetic algorithms are a viable computational

model of creative design

Page 8: Creative Design Using Collaborative Interactive Genetic Algorithms

Outline

1. Creativity in Design

2. Collaborative Interactive Genetic Algorithms

3. Reducing User Fatigue in Interactive Genetic Algorithms

4. Testing Our Computational Model of Creative Design

5. Contributions

Page 9: Creative Design Using Collaborative Interactive Genetic Algorithms

Collaborative Interactive Genetic Algorithm

• Population based search technique– Natural selection– Survival of the fittest

Page 10: Creative Design Using Collaborative Interactive Genetic Algorithms

CollaborativeInteractive Genetic Algorithms (IGAs)

• Fashion design (Kim 2000)• Micromachine design (Kamalian 2005)• Music, editorial design (Takagi 2001)• Traveling salesman problem (Louis 1999)

Page 11: Creative Design Using Collaborative Interactive Genetic Algorithms

Collaborative Interactive Genetic Algorithm

Page 12: Creative Design Using Collaborative Interactive Genetic Algorithms

Creative Design

• Floorplans with rectangular rooms• Purposely shifting the focus of

the search space– Circular rooms– Ellipsoid rooms– Star-shaped rooms

Page 13: Creative Design Using Collaborative Interactive Genetic Algorithms

Creative Design

Page 14: Creative Design Using Collaborative Interactive Genetic Algorithms

Sharing Solutions

Page 15: Creative Design Using Collaborative Interactive Genetic Algorithms

IGAP: Interactive Genetic Algorithm Peer to Peer

Page 16: Creative Design Using Collaborative Interactive Genetic Algorithms

Outline

1. Creativity in Design

2. Collaborative Interactive Genetic Algorithms

3. Reducing User Fatigue in Interactive Genetic Algorithms

4. Testing Our Computational Model of Creative Design

5. Contributions

Page 17: Creative Design Using Collaborative Interactive Genetic Algorithms

User Fatigue in Interactive Genetic Algorithms

• Genetic Algorithms tend to rely on– Large populations– Many generations

• Suboptimal solutions• Noisy fitness

Page 18: Creative Design Using Collaborative Interactive Genetic Algorithms

Fitness Interpolation

• Pick the best solution every nth generation

Page 19: Creative Design Using Collaborative Interactive Genetic Algorithms

Experimental Setup

• Test on the onemax problem• Subset methods– Best n, best n/2 and worst n/2, random n, PCA n

• Subset size• Gaussian noise• Collaboration

Page 20: Creative Design Using Collaborative Interactive Genetic Algorithms

Experimental Setup

• Simulated user input– 20 user evaluations– Greedy user always picks the solution with most

ones• 30 independent runs• Step sizes of 1, 2, 5• Subset size 9

Page 21: Creative Design Using Collaborative Interactive Genetic Algorithms

Boxplots

Maximum

Minimum

Median

Upper quartile

Lower quartile

Outlier

Outlier

Page 22: Creative Design Using Collaborative Interactive Genetic Algorithms

Subset Methods

Page 23: Creative Design Using Collaborative Interactive Genetic Algorithms

Subset Size

Step size 1Step size 2

Step size 5

Page 24: Creative Design Using Collaborative Interactive Genetic Algorithms

Subset Size

Page 25: Creative Design Using Collaborative Interactive Genetic Algorithms

No Noise vs Gaussian Noise with Sigma=1

Step size 1Step size 2

Step size 5

Page 26: Creative Design Using Collaborative Interactive Genetic Algorithms

Noise

Page 27: Creative Design Using Collaborative Interactive Genetic Algorithms

Number of Peers

Page 28: Creative Design Using Collaborative Interactive Genetic Algorithms

Summary

• Users can effectively bias evolution towards high fitness solutions– Subset size– Noise– Collaboration

Page 29: Creative Design Using Collaborative Interactive Genetic Algorithms

Outline

1. Creativity in Design

2. Collaborative Interactive Genetic Algorithms

3. Reducing User Fatigue in Interactive Genetic Algorithms

4. Testing Our Computational Model of Creative Design

5. Contributions

Page 30: Creative Design Using Collaborative Interactive Genetic Algorithms

Goals

• User studies– Solutions created individually– Solutions created collaboratively

• Show that solutions created collaboratively are more creative

Page 31: Creative Design Using Collaborative Interactive Genetic Algorithms

First User Study: Floorplanning

7 8 9 10 11 12

Bedroom

Living RoomEating area

Bathroom

Page 32: Creative Design Using Collaborative Interactive Genetic Algorithms

Collaborative Floorplanning

User’s Individuals Peers’ Individuals

Page 33: Creative Design Using Collaborative Interactive Genetic Algorithms

Pilot: Experimental Setup

• Requirements– Design a floorplan for a 2 bedroom, 1 bathroom

apartment– Living room should face north-west– The two bedrooms should not have a common

wall– At least one of the bedrooms should have direct

access to the bathroom

1 2 3 4 5 6

Page 34: Creative Design Using Collaborative Interactive Genetic Algorithms

Pilot: Experimental Setup

• Four colleagues and I evolved floorplans– Individually– Collaboratively

• Ten computer science graduate students evaluated the designs by taking a survey

• The plans were evaluated for creative content based on practicality and originality

1 2 3 4 5 6

Page 35: Creative Design Using Collaborative Interactive Genetic Algorithms

Floorplan Results

1 2 3 4 5 6

7 8 9 10 11 12

Page 36: Creative Design Using Collaborative Interactive Genetic Algorithms

ResultsDESIGN # PRACTICALITY ORIGINALITY RANK

1 2.7 3.1 6 2 2.9 2.7 8 3 2.4 3.3 7 4 4.4 3.3 2 6 4.0 3.2 3 7 2.2 3.4 8 8 2.8 3.4 5 10 3.2 3.5 4 11 4.2 3.6 1 12 1.6 3.8 10

7 8 9 10 11 12

1 2 3 4 5 6

Page 37: Creative Design Using Collaborative Interactive Genetic Algorithms

Floorplanning User Study: Experimental Setup

• Requirements– Create a floorplan for a 2 bedroom, 1 bathroom

apartment– Bathrooms close to the bedrooms– Bathrooms far from kitchen and dining areas

Page 38: Creative Design Using Collaborative Interactive Genetic Algorithms

Floorplanning User Study: Experimental Setup

• Participants:– 8 women, 12 men

• Five groups of size four• Agenda

1. Tutorial2. Create individual floorplan3. Create collaborative floorplan4. Evaluation of floorplans

Page 39: Creative Design Using Collaborative Interactive Genetic Algorithms

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

Page 40: Creative Design Using Collaborative Interactive Genetic Algorithms

Hypothesis

• Is collaboration amongst peers sufficient to allow for the potential to produce creative solutions?

• Designs evolved collaboratively will consistently rank higher in the evaluation criteria.

Page 41: Creative Design Using Collaborative Interactive Genetic Algorithms

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

Page 42: Creative Design Using Collaborative Interactive Genetic Algorithms

Discussion

• Ambiguity in evaluation criteria– Appealing – unappealing– Positive – Negative (?)– Negative – Positive (?)

• Applicability of evaluation criteria– “Exciting”– Domain expert vs. student

• Participants created only 1 collaborative floorplan and 1 individual floorplan

• Simple graphic representation

Page 43: Creative Design Using Collaborative Interactive Genetic Algorithms

Second User Study: 3D Modeling

• Vertex programs– p.x += 20– p.x += sin(time)

Page 44: Creative Design Using Collaborative Interactive Genetic Algorithms

Sample Ninja Transformations

Page 45: Creative Design Using Collaborative Interactive Genetic Algorithms

Collaborative Setup

• User 1– Equations that modify the

x and z coordinates

• User 2– Equations that modify the

y and z coordinates

• After collaboration– Equations that modify the

x, y, and z coordinates

Page 46: Creative Design Using Collaborative Interactive Genetic Algorithms

Experimental Setup

• Design Phase– Two groups of 10 participants

• Evaluation Phase– On-site evaluation• 20 participants

– Online evaluation• 16 participants

Page 47: Creative Design Using Collaborative Interactive Genetic Algorithms

Experimental Setup

• Groups of 2• Agenda

1. Tutorial2. Creating 3D models3. Picking solutions for

the evaluation phase

Individually

Collaboratively

Individually

Collaboratively

Individually

Collaboratively

Page 48: Creative Design Using Collaborative Interactive Genetic Algorithms

Design Phase

Female

Ninja

Robot

Robot

Ninja

Female

Individually

Collaboratively

Individually

Collaboratively

Individually

Collaboratively

Female

Ninja

Robot

Female

Ninja

Robot

Page 49: Creative Design Using Collaborative Interactive Genetic Algorithms

Evaluation Phase

• 7 point Likert scale• Creative Product Semantic Scale• The transformation is:

– Extremely creative – Not Creative At All

• The transformation can be used in a video game.

• The transformation with minor tweaks can be used in a video game.

• The transformation is novel.• The transformation is surprising.

Individually

Collaboratively

Individually

Collaboratively

Individually

Collaboratively

Page 50: Creative Design Using Collaborative Interactive Genetic Algorithms

Results

Individual Collaborative

Page 51: Creative Design Using Collaborative Interactive Genetic Algorithms

The transformation is creative.

Page 52: Creative Design Using Collaborative Interactive Genetic Algorithms

The transformation can be used in a video game.

Page 53: Creative Design Using Collaborative Interactive Genetic Algorithms

The transformation with minor tweaks can be used in a video game.

Page 54: Creative Design Using Collaborative Interactive Genetic Algorithms

The transformation is novel.

Page 55: Creative Design Using Collaborative Interactive Genetic Algorithms

The transformation is surprising.

Page 56: Creative Design Using Collaborative Interactive Genetic Algorithms

Online Evaluation

• Best individually created models• Best collaboratively created models• Evaluation Criteria– The transformation is creative.– The transformation can be used in a video game.– The transformation is novel.– The transformation is surprising.– Which of the two rows did you like the most?– Which of the two rows is the most creative?

Page 57: Creative Design Using Collaborative Interactive Genetic Algorithms

Results

• Individually created models vs collaboratively created models– No statistically significant results

Page 58: Creative Design Using Collaborative Interactive Genetic Algorithms

Results

• Which of the two rows did you like the most?– 8 participants picked the

individual row– 7 participants picked the

collaborative row– 1 participant did not answer

• Which of the two rows is the most creative?– 3 participants picked the

individual row– 13 participants picked the

collaborative row

IndividualCollaborativeNA

IndividualCollaborative

Page 59: Creative Design Using Collaborative Interactive Genetic Algorithms

Discussion

• Different 3D models• Lack of context• Online Evaluation Nuances– Switching windows– 15 second average– Rewinding– Scoring the row of individually created models first– Indecisive participants and median scores

Page 60: Creative Design Using Collaborative Interactive Genetic Algorithms

Outline

1. Creativity in Design

2. Collaborative Interactive Genetic Algorithms

3. Reducing User Fatigue in Interactive Genetic Algorithms

4. Testing Our Computational Model of Creative Design

5. Contributions

Page 61: Creative Design Using Collaborative Interactive Genetic Algorithms

Contributions

• A new computational model of creative design– Subjective exploration of solutions– Integrates collaboration

• Implementation of IGAP framework: Interactive Genetic Algorithm Peer to Peer

• Analysis of our fitness interpolation technique in the onemax problem

Page 62: Creative Design Using Collaborative Interactive Genetic Algorithms

Contributions

• Floorplanning pilot– Collaborative solutions were considered more

original• Floorplanning user study– Collaborative solutions were considered more

original and revolutionary• 3D Modeling user study– 13 out of 16 participants picked row of

collaborative of solutions as the most creative

Page 63: Creative Design Using Collaborative Interactive Genetic Algorithms

Future Work

• Conduct additional user studies– Long term user studies with design teams

• Refine and test IGAP framework• Machine learning

Page 64: Creative Design Using Collaborative Interactive Genetic Algorithms

Acknowledgments

• Dr. Sushil Louis• Dr. Bobby Bryant• Dr. Swatee Naik• Dr. Sergiu Dascalu• Dr. Amit Banerjee• Dr. Darren Platt• Study participants

– Students, adult volunteers, and faculty• This work was supported in part by contract number N00014-

05-1-0709 from the Office of Naval Research and the National Science Foundation under Grant no. 0447416.

Page 65: Creative Design Using Collaborative Interactive Genetic Algorithms

Questions?

Thank you!

www.cse.unr.edu/~quiroz