Presentation | 2018-06-13 Enumeration of Distinct Support Vectors for Model Selection Kentaro Kanamori, Satoshi Hara, Masakazu Ishihata, Hiroki Arimura, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In ordinary machine learning problems, the learning algorithm outputs a single model that optimizes its learning objective function, and doesn't consider the other models. In contrast, recently, some methods for enumerating multiple models are presented. For example, Hara and Maehara presented an algorithm for Lasso (AAAI'17), while, Ruggieri for decision tree learning (ICML'17). In this paper, we extend the algorithm of Hara and Maehara, we present an efficient algorithm for the $K$-best model enumeration problem in the training of Support Vector Machines (SVM). This algorithm is, given a non-negative number $K>0$, enumerating $K$-best models that have distinct support vectors in the descending order of the values of the objective function of SVM. By experiments on synthetic and real datasets, we evaluate the efficiency and effectiveness of our algorithm. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Support Vector Machine (SVM) / model enumeration / model selection / subsample selection |
Paper # | IBISML2018-12 |
Date of Issue | 2018-06-06 (IBISML) |
Conference Information | |
Committee | NC / IBISML / IPSJ-BIO / IPSJ-MPS |
---|---|
Conference Date | 2018/6/13(3days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Okinawa Institute of Science and Technology |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Machine Learning Approach to Biodata Mining, and General |
Chair | Yutaka Hirata(Chubu Univ.) / Hisashi Kashima(Kyoto Univ.) |
Vice Chair | Hayaru Shouno(UEC) / Masashi Sugiyama(Univ. of Tokyo) / Koji Tsuda(Univ. of Tokyo) |
Secretary | Hayaru Shouno(Nagoya Univ.) / Masashi Sugiyama(NAIST) / Koji Tsuda(Nagoya Inst. of Tech.) / (AIST) |
Assistant | Keiichiro Inagaki(Chubu Univ.) / Takashi Shinozaki(NICT) / Tomoharu Iwata(NTT) / Shigeyuki Oba(Kyoto Univ.) |
Paper Information | |
Registration To | Technical Committee on Neurocomputing / Technical Committee on Infomation-Based Induction Sciences and Machine Learning / Special Interest Group on Bioinformatics and Genomics / Special Interest Group on Mathematical Modeling and Problem Solving |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Enumeration of Distinct Support Vectors for Model Selection |
Sub Title (in English) | |
Keyword(1) | Support Vector Machine (SVM) |
Keyword(2) | model enumeration |
Keyword(3) | model selection |
Keyword(4) | subsample selection |
1st Author's Name | Kentaro Kanamori |
1st Author's Affiliation | Hokkaido University(Hokaido Univ.) |
2nd Author's Name | Satoshi Hara |
2nd Author's Affiliation | Osaka University(Osaka Univ.) |
3rd Author's Name | Masakazu Ishihata |
3rd Author's Affiliation | NTT Communication Science Laboratories(NTT) |
4th Author's Name | Hiroki Arimura |
4th Author's Affiliation | Hokkaido University(Hokaido Univ.) |
Date | 2018-06-13 |
Paper # | IBISML2018-12 |
Volume (vol) | vol.118 |
Number (no) | IBISML-81 |
Page | pp.pp.81-88(IBISML), |
#Pages | 8 |
Date of Issue | 2018-06-06 (IBISML) |