A deep learning approach to multiple kernel fusion

Huan Song, Jayaraman J. Thiagarajan, Prasanna Sattigeri, Karthikeyan Natesan Ramamurthy, Andreas Spanias

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

Kernel fusion is a popular and effective approach for combining multiple features that characterize different aspects of data. Traditional approaches for Multiple Kernel Learning (MKL) attempt to learn the parameters for combining the kernels through sophisticated optimization procedures. In this paper, we propose an alternative approach that creates dense embeddings for data using the kernel similarities and adopts a deep neural network architecture for fusing the embeddings. In order to improve the effectiveness of this network, we introduce the kernel dropout regularization strategy coupled with the use of an expanded set of composition kernels. Experiment results on a real-world activity recognition dataset show that the proposed architecture is effective in fusing kernels and achieves state-of-the-art performance.

Original languageEnglish (US)
Title of host publication2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages2292-2296
Number of pages5
ISBN (Electronic)9781509041176
DOIs
StatePublished - Jun 16 2017
Event2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - New Orleans, United States
Duration: Mar 5 2017Mar 9 2017

Other

Other2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017
CountryUnited States
CityNew Orleans
Period3/5/173/9/17

Fingerprint

Network architecture
Fusion reactions
Chemical analysis
Experiments
Deep neural networks
Deep learning

Keywords

  • Activity recognition
  • Deep learning
  • Dropout regularization
  • Kernel fusion

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Song, H., Thiagarajan, J. J., Sattigeri, P., Ramamurthy, K. N., & Spanias, A. (2017). A deep learning approach to multiple kernel fusion. In 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings (pp. 2292-2296). [7952565] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2017.7952565

A deep learning approach to multiple kernel fusion. / Song, Huan; Thiagarajan, Jayaraman J.; Sattigeri, Prasanna; Ramamurthy, Karthikeyan Natesan; Spanias, Andreas.

2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. p. 2292-2296 7952565.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Song, H, Thiagarajan, JJ, Sattigeri, P, Ramamurthy, KN & Spanias, A 2017, A deep learning approach to multiple kernel fusion. in 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings., 7952565, Institute of Electrical and Electronics Engineers Inc., pp. 2292-2296, 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017, New Orleans, United States, 3/5/17. https://doi.org/10.1109/ICASSP.2017.7952565
Song H, Thiagarajan JJ, Sattigeri P, Ramamurthy KN, Spanias A. A deep learning approach to multiple kernel fusion. In 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2017. p. 2292-2296. 7952565 https://doi.org/10.1109/ICASSP.2017.7952565
Song, Huan ; Thiagarajan, Jayaraman J. ; Sattigeri, Prasanna ; Ramamurthy, Karthikeyan Natesan ; Spanias, Andreas. / A deep learning approach to multiple kernel fusion. 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 2292-2296
@inproceedings{61494b12a83844f9ae70c4d64dbd228a,
title = "A deep learning approach to multiple kernel fusion",
abstract = "Kernel fusion is a popular and effective approach for combining multiple features that characterize different aspects of data. Traditional approaches for Multiple Kernel Learning (MKL) attempt to learn the parameters for combining the kernels through sophisticated optimization procedures. In this paper, we propose an alternative approach that creates dense embeddings for data using the kernel similarities and adopts a deep neural network architecture for fusing the embeddings. In order to improve the effectiveness of this network, we introduce the kernel dropout regularization strategy coupled with the use of an expanded set of composition kernels. Experiment results on a real-world activity recognition dataset show that the proposed architecture is effective in fusing kernels and achieves state-of-the-art performance.",
keywords = "Activity recognition, Deep learning, Dropout regularization, Kernel fusion",
author = "Huan Song and Thiagarajan, {Jayaraman J.} and Prasanna Sattigeri and Ramamurthy, {Karthikeyan Natesan} and Andreas Spanias",
year = "2017",
month = "6",
day = "16",
doi = "10.1109/ICASSP.2017.7952565",
language = "English (US)",
pages = "2292--2296",
booktitle = "2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - GEN

T1 - A deep learning approach to multiple kernel fusion

AU - Song, Huan

AU - Thiagarajan, Jayaraman J.

AU - Sattigeri, Prasanna

AU - Ramamurthy, Karthikeyan Natesan

AU - Spanias, Andreas

PY - 2017/6/16

Y1 - 2017/6/16

N2 - Kernel fusion is a popular and effective approach for combining multiple features that characterize different aspects of data. Traditional approaches for Multiple Kernel Learning (MKL) attempt to learn the parameters for combining the kernels through sophisticated optimization procedures. In this paper, we propose an alternative approach that creates dense embeddings for data using the kernel similarities and adopts a deep neural network architecture for fusing the embeddings. In order to improve the effectiveness of this network, we introduce the kernel dropout regularization strategy coupled with the use of an expanded set of composition kernels. Experiment results on a real-world activity recognition dataset show that the proposed architecture is effective in fusing kernels and achieves state-of-the-art performance.

AB - Kernel fusion is a popular and effective approach for combining multiple features that characterize different aspects of data. Traditional approaches for Multiple Kernel Learning (MKL) attempt to learn the parameters for combining the kernels through sophisticated optimization procedures. In this paper, we propose an alternative approach that creates dense embeddings for data using the kernel similarities and adopts a deep neural network architecture for fusing the embeddings. In order to improve the effectiveness of this network, we introduce the kernel dropout regularization strategy coupled with the use of an expanded set of composition kernels. Experiment results on a real-world activity recognition dataset show that the proposed architecture is effective in fusing kernels and achieves state-of-the-art performance.

KW - Activity recognition

KW - Deep learning

KW - Dropout regularization

KW - Kernel fusion

UR - http://www.scopus.com/inward/record.url?scp=85023770195&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85023770195&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2017.7952565

DO - 10.1109/ICASSP.2017.7952565

M3 - Conference contribution

SP - 2292

EP - 2296

BT - 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -