Bilevel model-based discriminative dictionary learning for recognition

Most supervised dictionary learning methods optimize the combinations of reconstruction error, sparsity prior, and discriminative terms. Thus, the learnt dictionaries may not be optimal for recognition tasks. Also, the sparse codes learning models in the training and the testing phases are inconsist...

Full description

Saved in:
Bibliographic Details
Main Authors: ZHOU, Pan, ZHANG, Chao, LIN Zhouchen
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2016
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/8967
https://ink.library.smu.edu.sg/context/sis_research/article/9970/viewcontent/2016_TIP_bilevel.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
id sg-smu-ink.sis_research-9970
record_format dspace
spelling sg-smu-ink.sis_research-99702024-07-17T06:52:27Z Bilevel model-based discriminative dictionary learning for recognition ZHOU, Pan ZHANG, Chao LIN Zhouchen, Most supervised dictionary learning methods optimize the combinations of reconstruction error, sparsity prior, and discriminative terms. Thus, the learnt dictionaries may not be optimal for recognition tasks. Also, the sparse codes learning models in the training and the testing phases are inconsistent. Besides, without utilizing the intrinsic data structure, many dictionary learning methods only employ the 0 or 1 norm to encode each datum independently, limiting the performance of the learnt dictionaries. We present a novel bilevel model-based discriminative dictionary learning method for recognition tasks. The upper level directly minimizes the classification error, while the lower level uses the sparsity term and the Laplacian term to characterize the intrinsic data structure. The lower level is subordinate to the upper level. Therefore, our model achieves an overall optimality for recognition in that the learnt dictionary is directly tailored for recognition. Moreover, the sparse codes learning models in the training and the testing phases can be the same. We further propose a novel method to solve our bilevel optimization problem. It first replaces the lower level with its Karush–Kuhn–Tucker conditions and then applies the alternating direction method of multipliers to solve the equivalent problem. Extensive experiments demonstrate the effectiveness and robustness of our method. 2016-10-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/8967 info:doi/10.1109/TIP.2016.2623487 https://ink.library.smu.edu.sg/context/sis_research/article/9970/viewcontent/2016_TIP_bilevel.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Sparse representation dictionary learning bilevel optimization recognition alternating direction method Artificial Intelligence and Robotics Software Engineering
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic Sparse representation
dictionary learning
bilevel optimization
recognition
alternating direction method
Artificial Intelligence and Robotics
Software Engineering
spellingShingle Sparse representation
dictionary learning
bilevel optimization
recognition
alternating direction method
Artificial Intelligence and Robotics
Software Engineering
ZHOU, Pan
ZHANG, Chao
LIN Zhouchen,
Bilevel model-based discriminative dictionary learning for recognition
description Most supervised dictionary learning methods optimize the combinations of reconstruction error, sparsity prior, and discriminative terms. Thus, the learnt dictionaries may not be optimal for recognition tasks. Also, the sparse codes learning models in the training and the testing phases are inconsistent. Besides, without utilizing the intrinsic data structure, many dictionary learning methods only employ the 0 or 1 norm to encode each datum independently, limiting the performance of the learnt dictionaries. We present a novel bilevel model-based discriminative dictionary learning method for recognition tasks. The upper level directly minimizes the classification error, while the lower level uses the sparsity term and the Laplacian term to characterize the intrinsic data structure. The lower level is subordinate to the upper level. Therefore, our model achieves an overall optimality for recognition in that the learnt dictionary is directly tailored for recognition. Moreover, the sparse codes learning models in the training and the testing phases can be the same. We further propose a novel method to solve our bilevel optimization problem. It first replaces the lower level with its Karush–Kuhn–Tucker conditions and then applies the alternating direction method of multipliers to solve the equivalent problem. Extensive experiments demonstrate the effectiveness and robustness of our method.
format text
author ZHOU, Pan
ZHANG, Chao
LIN Zhouchen,
author_facet ZHOU, Pan
ZHANG, Chao
LIN Zhouchen,
author_sort ZHOU, Pan
title Bilevel model-based discriminative dictionary learning for recognition
title_short Bilevel model-based discriminative dictionary learning for recognition
title_full Bilevel model-based discriminative dictionary learning for recognition
title_fullStr Bilevel model-based discriminative dictionary learning for recognition
title_full_unstemmed Bilevel model-based discriminative dictionary learning for recognition
title_sort bilevel model-based discriminative dictionary learning for recognition
publisher Institutional Knowledge at Singapore Management University
publishDate 2016
url https://ink.library.smu.edu.sg/sis_research/8967
https://ink.library.smu.edu.sg/context/sis_research/article/9970/viewcontent/2016_TIP_bilevel.pdf
_version_ 1814047660258099200