Presentation | 2019-03-14 Convergence-guaranteed independent positive semidefinite tensor analysis for blind source separation Kanta Fukushige, Norihiro Takamune, Daichi Kitamura, Hiroshi Saruwatari, Rintaro Ikeshita, Tomohiro Nakatani, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | This paper focuses on independent positive semidefinite tensor analysis (IPSDTA), which is a technique for over-determined blind source separation. IPSDTA alternately updates demixing matrices and source models for minimizing the cost function. However, theoretical convergence of the conventional optimization algorithm has not been discussed. In this paper, we point out that the update step of the demixing matrix in the conventional algorithm lacks the theoretical convergence. Also, a new convergence-guaranteed optimization algorithm based on vectorwise coordinate descent is proposed. Experimental results show the validity of proposed IPSDTA regarding both the convergence and the performance of source separation. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Blind source separation / Positive semidefinite tensor factorization / Independent positive semidefinite tensor analysis / Convergence guarantee |
Paper # | EA2018-127,SIP2018-133,SP2018-89 |
Date of Issue | 2019-03-07 (EA, SIP, SP) |
Conference Information | |
Committee | EA / SIP / SP |
---|---|
Conference Date | 2019/3/14(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | i+Land nagasaki (Nagasaki-shi) |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Engineering/Electro Acoustics, Signal Processing, Speech, and Related Topics |
Chair | Suehiro Shimauchi(Kanazawa Inst. of Tech.) / Shogo Muramatsu(Niigata Univ.) / Yoichi Yamashita(Ritsumeikan Univ.) |
Vice Chair | Kenichi Furuya(Oita Univ.) / Kanji Watanabe(Akita Pref. Univ.) / Naoyuki Aikawa(TUS) / Kazunori Hayashi(Osaka City Univ) / Akinobu Ri(Nagoya Inst. of Tech.) |
Secretary | Kenichi Furuya(Shizuoka Inst. of Science and Tech.) / Kanji Watanabe(NHK) / Naoyuki Aikawa(Takushoku Univ.) / Kazunori Hayashi(Hiroshima Univ.) / Akinobu Ri(Kyoto Univ.) |
Assistant | Keisuke Imoto(Ritsumeikan Univ.) / Daisuke Morikawa(Toyama Pref Univ.) / Katsumi Konishi(Hosei Univ.) / hyihsin(Takushoku Univ.) / Tomoki Koriyama(Tokyo Inst. of Tech.) / Satoshi Kobashikawa(NTT) |
Paper Information | |
Registration To | Technical Committee on Engineering Acoustics / Technical Committee on Signal Processing / Technical Committee on Speech |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Convergence-guaranteed independent positive semidefinite tensor analysis for blind source separation |
Sub Title (in English) | |
Keyword(1) | Blind source separation |
Keyword(2) | Positive semidefinite tensor factorization |
Keyword(3) | Independent positive semidefinite tensor analysis |
Keyword(4) | Convergence guarantee |
1st Author's Name | Kanta Fukushige |
1st Author's Affiliation | The University of Tokyo(UTokyo) |
2nd Author's Name | Norihiro Takamune |
2nd Author's Affiliation | The University of Tokyo(UTokyo) |
3rd Author's Name | Daichi Kitamura |
3rd Author's Affiliation | National Institute of Technology, Kagawa College(Kagawa-NICT) |
4th Author's Name | Hiroshi Saruwatari |
4th Author's Affiliation | The University of Tokyo(UTokyo) |
5th Author's Name | Rintaro Ikeshita |
5th Author's Affiliation | NTT Communication Science Laboratories(NTT) |
6th Author's Name | Tomohiro Nakatani |
6th Author's Affiliation | NTT Communication Science Laboratories(NTT) |
Date | 2019-03-14 |
Paper # | EA2018-127,SIP2018-133,SP2018-89 |
Volume (vol) | vol.118 |
Number (no) | EA-495,SIP-496,SP-497 |
Page | pp.pp.167-172(EA), pp.167-172(SIP), pp.167-172(SP), |
#Pages | 6 |
Date of Issue | 2019-03-07 (EA, SIP, SP) |