Presentation | 2019-05-17 An investigation of the most "kawaii" shape using preferential Bayesian optimization Masashi Komoei, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | This study aims to explore a psychophysical latent function of perceived “Kawaii” (cuteness) against multidimensional shape parameters using a Bayesian optimization (BO) methodology. BO is an effective approach to sequentially optimize the black-box function in cases where the cost for evaluations are high. In standard BO applications (e.g., parameter tuning for machine learning), it is possible to query objective functions directly. However, humans are better at evaluating differences rather than absolute magnitudes. Thus, instead of standard BO, we performed preferential BO, which enables the estimation of the latent function through two-alternative forced choice task. To generate the pairs of stimulus images for the comparison, we used the Expected Improvement (EI) acquisition function. Each contour shape stimulus was created by elliptic Fourier descriptors (EFDs) based on six parameters (up to the third harmonic) calculated by Gaussian process regression and EI. Fourteen participants provided their preferences for the pairs of contour images in seventy trials. Based on the results, we estimated the most “kawaii” shape and the least “kawaii” shape on average. The most kawaii shape was found to be round and to have two protrusions on the top. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Bayesian Optimization / Kawaii / Shape |
Paper # | HCS2019-15,HIP2019-15 |
Date of Issue | 2019-05-09 (HCS, HIP) |
Conference Information | |
Committee | HCS / HIP / HI-SIGCE |
---|---|
Conference Date | 2019/5/16(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Okinawa Industry Support Center |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Masafumi Matsuda(NTT) / Miyuki Kamachi(Kogakuin Univ.) |
Vice Chair | Nobuyuki Watanabe(Kanazawa Inst. of Tech.) / Tomoo Inoue(Univ. of Tsukuba) / Shuichi Sakamoto(Tohoku Univ.) / Yuji Wada(Ritsumeikan Univ.) |
Secretary | Nobuyuki Watanabe(Ritsumeikan Univ.) / Tomoo Inoue(Osaka Electro-Comm. Univ.) / Shuichi Sakamoto(NEC) / Yuji Wada(Univ. of Tsukuba) / (NICT) |
Assistant | Kazuki Takashima(Tohoku Univ.) / Ken Fujiwara(Osaka Univ. of Economic) / Kazunori Terada(Gifu Univ.) / Atsushi Kimura(Nihon Univ.) / Atsushi Wada(NICT) / Hidetoshi Kanaya(Ritsumeikan Univ.) / Yuki Yamada(Kyushu Univ.) |
Paper Information | |
Registration To | Technical Committee on Human Communication Science / Technical Committee on Human Information Processing / Special Interest Group on Communication Enhancement |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | An investigation of the most "kawaii" shape using preferential Bayesian optimization |
Sub Title (in English) | |
Keyword(1) | Bayesian Optimization |
Keyword(2) | Kawaii |
Keyword(3) | Shape |
1st Author's Name | Masashi Komoei |
1st Author's Affiliation | Osaka Electro-Communication University(OECU) |
Date | 2019-05-17 |
Paper # | HCS2019-15,HIP2019-15 |
Volume (vol) | vol.119 |
Number (no) | HCS-38,HIP-39 |
Page | pp.pp.109-112(HCS), pp.109-112(HIP), |
#Pages | 4 |
Date of Issue | 2019-05-09 (HCS, HIP) |