Presentation | 2017-03-01 [Invited Talk] Multikernel Adaptive Filtering: Signal Processing and Machine Learning Masahiro Yukawa, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | We present the multikernel adaptive filtering and introduce its recent advances. Multikernel adaptive filtering is a recently proposed learning paradigm for nonlinear estimation tasks, and its potential applications include time-series data prediction and multi-scale data analysis, among many others. Kernel adaptive filter is an online extension of the popular Gaussian process or support vector regression, and its performance depends significantly on the selected kernel. Multikernel adaptive filtering has been proposed to circumvent this notoriously difficult kernel-dependency problem which is a common issue for most (if not all) kernel methods. Kernel adaptive filtering algorithm can be constructed based on the convex projection in one of the following spaces: the parameter space or the reproducing kernel Hilbert space (RKHS). While it was shown experimentally that the use of the latter space leads to faster convergence, its theoretical reason becomes clear only recently. The multikernel adaptive filtering algorithm based on the convex projection in the product space of RKHSs gives improved convergence behaviors compared to the original algorithms, which are based on the Euclidean-space projection. We also touch upon two types of multi scale, an application to communication systems, and comparisons with multiple kernel learning, which has been studied in machine learning community. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | nonlinear adaptive filter / time-series data prediction / reproducing kernel Hilbert space / convex projection |
Paper # | EA2016-113,SIP2016-168,SP2016-108 |
Date of Issue | 2017-02-22 (EA, SIP, SP) |
Conference Information | |
Committee | SP / SIP / EA |
---|---|
Conference Date | 2017/3/1(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Okinawa Industry Support Center |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Speech, Engineering/Electro Acoustics, Signal Processing, and Related Topics |
Chair | Kazunori Mano(Shibaura Inst. of Tech.) / Makoto Nakashizuka(Chiba Inst. of Tech.) / Mitsunori Mizumachi(Kyushu Inst. of Tech.) |
Vice Chair | Hiroki Mori(Utsunomiya Univ.) / Masahiro Okuda(Univ. of Kitakyushu) / Shogo Muramatsu(Niigata Univ.) / Yoichi Haneda(Univ. of Electro-Comm.) / Suehiro Shimauchi(NTT) |
Secretary | Hiroki Mori(Kobe Univ.) / Masahiro Okuda(Shizuoka Univ.) / Shogo Muramatsu(Ritsumeikan Univ.) / Yoichi Haneda(Chiba Inst. of Tech.) / Suehiro Shimauchi(KDDI R&D Labs.) |
Assistant | Taichi Asami(NTT) / Kei Hashimoto(Nagoya Inst. of Tech.) / Osamu Watanabe(Takushoku Univ.) / Shigeto Takeoka(Shizuoka Inst. of Science and Tech.) / TREVINO Jorge(Tohoku Univ.) |
Paper Information | |
Registration To | Technical Committee on Speech / Technical Committee on Signal Processing / Technical Committee on Engineering Acoustics |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | [Invited Talk] Multikernel Adaptive Filtering: Signal Processing and Machine Learning |
Sub Title (in English) | |
Keyword(1) | nonlinear adaptive filter |
Keyword(2) | time-series data prediction |
Keyword(3) | reproducing kernel Hilbert space |
Keyword(4) | convex projection |
1st Author's Name | Masahiro Yukawa |
1st Author's Affiliation | Keio University(Keio Univ.) |
Date | 2017-03-01 |
Paper # | EA2016-113,SIP2016-168,SP2016-108 |
Volume (vol) | vol.116 |
Number (no) | EA-475,SIP-476,SP-477 |
Page | pp.pp.177-182(EA), pp.177-182(SIP), pp.177-182(SP), |
#Pages | 6 |
Date of Issue | 2017-02-22 (EA, SIP, SP) |