IEICE Technical Committee Submission System
Conference Paper's Information
Online Proceedings
[Sign in]
... (for ESS/CS/ES/ISS)
Tech. Rep. Archives
... (for ES/CS)
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2017-03-01 15:55
[Invited Talk] Multikernel Adaptive Filtering: Signal Processing and Machine Learning
Masahiro Yukawa (Keio Univ.)
Abstract (in Japanese) (See Japanese page) 
(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) 
(in English) nonlinear adaptive filter / time-series data prediction / reproducing kernel Hilbert space / convex projection / / / /  
Reference Info. IEICE Tech. Rep., vol. 116, no. 477, SP2016-108, pp. 177-182, March 2017.
Paper # SP2016-108 
Date of Issue 2017-02-22 (EA, SIP, SP) 
ISSN Print edition: ISSN 0913-5685  Online edition: ISSN 2432-6380

Conference Information
Committee SP SIP EA  
Conference Date 2017-03-01 - 2017-03-02 
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 
Paper Information
Registration To SP 
Conference Code 2017-03-SP-SIP-EA 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) 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  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Masahiro Yukawa  
1st Author's Affiliation Keio University (Keio Univ.)
2nd Author's Name  
2nd Author's Affiliation ()
3rd Author's Name  
3rd Author's Affiliation ()
4th Author's Name  
4th Author's Affiliation ()
5th Author's Name  
5th Author's Affiliation ()
6th Author's Name  
6th Author's Affiliation ()
7th Author's Name  
7th Author's Affiliation ()
8th Author's Name  
8th Author's Affiliation ()
9th Author's Name  
9th Author's Affiliation ()
10th Author's Name  
10th Author's Affiliation ()
11th Author's Name  
11th Author's Affiliation ()
12th Author's Name  
12th Author's Affiliation ()
13th Author's Name  
13th Author's Affiliation ()
14th Author's Name  
14th Author's Affiliation ()
15th Author's Name  
15th Author's Affiliation ()
16th Author's Name  
16th Author's Affiliation ()
17th Author's Name  
17th Author's Affiliation ()
18th Author's Name  
18th Author's Affiliation ()
19th Author's Name  
19th Author's Affiliation ()
20th Author's Name  
20th Author's Affiliation ()
Speaker
Date Time 2017-03-01 15:55:00 
Presentation Time 40 
Registration for SP 
Paper # IEICE-EA2016-113,IEICE-SIP2016-168,IEICE-SP2016-108 
Volume (vol) IEICE-116 
Number (no) no.475(EA), no.476(SIP), no.477(SP) 
Page pp.177-182 
#Pages IEICE-6 
Date of Issue IEICE-EA-2017-02-22,IEICE-SIP-2017-02-22,IEICE-SP-2017-02-22 


[Return to Top Page]

[Return to IEICE Web Page]


The Institute of Electronics, Information and Communication Engineers (IEICE), Japan