TY - GEN
T1 - Computer-vision Classification-algorithms Are Inherently Creative When Error-prone
AU - Hoorn, Johan F.
N1 - Funding Information:
This study was supported by project P0000254 - AntiFix: Computational Creativity for Industrial Engineering, funding scheme: Start-up Fund for New Recruits, grant number 1-BE22. Ethical approval was obtained from the Ethical Review Board of the university, file number HSEARS20210923002. Desmond Germans is acknowledged for his implementation of the ACASIA simulator.
Publisher Copyright:
© 2022 ACM.
PY - 2022/12/27
Y1 - 2022/12/27
N2 - Whether coming from a linear support vector machine, from logistic regression, or a quasi-Newtonian, the fine-tuning of the decision boundary in any given data set is essential to mitigate the loss term so that neural nets in image recognition can divide a data space into separate sections and correctly classify an input. By their very nature, neural nets are logically non-deterministic but rest on probability-weighted associations, which are adjusted recursively to enhance the similarity of intermediate results to the target output, the remaining difference being the error.' However, taxonomies should not be crisp but seen as fuzzy classes, allowing for hybrid exemplars that transgress category boundaries. The associative and similarity orientation of neural nets and deep learning makes such systems inherently creative in that misclassifications are at the basis of creative crossovers in information processing. This new conceptualization of network errors is supported by the ratings of 40 top-ranking designers of 20 image-recognition mistakes on the dimensions of creativity and innovativeness.
AB - Whether coming from a linear support vector machine, from logistic regression, or a quasi-Newtonian, the fine-tuning of the decision boundary in any given data set is essential to mitigate the loss term so that neural nets in image recognition can divide a data space into separate sections and correctly classify an input. By their very nature, neural nets are logically non-deterministic but rest on probability-weighted associations, which are adjusted recursively to enhance the similarity of intermediate results to the target output, the remaining difference being the error.' However, taxonomies should not be crisp but seen as fuzzy classes, allowing for hybrid exemplars that transgress category boundaries. The associative and similarity orientation of neural nets and deep learning makes such systems inherently creative in that misclassifications are at the basis of creative crossovers in information processing. This new conceptualization of network errors is supported by the ratings of 40 top-ranking designers of 20 image-recognition mistakes on the dimensions of creativity and innovativeness.
KW - Computational creativity
KW - Deep learning
KW - Error
KW - Misclassification
KW - Neural networks
UR - http://www.scopus.com/inward/record.url?scp=85147250093&partnerID=8YFLogxK
U2 - 10.1145/3574131.3574444
DO - 10.1145/3574131.3574444
M3 - Conference article published in proceeding or book
AN - SCOPUS:85147250093
T3 - Proceedings - VRCAI 2022: 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry
BT - Proceedings - VRCAI 2022
A2 - Spencer, Stephen N.
PB - Association for Computing Machinery, Inc
T2 - 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry, VRCAI 2022
Y2 - 27 December 2022 through 29 December 2022
ER -