An information bottleneck approach to optimize the dictionary of visual data

Wilson, S and C, Krishna Mohan (2017) An information bottleneck approach to optimize the dictionary of visual data. IEEE Transactions on Multimedia. p. 1. ISSN 1520-9210 (In Press)

Full text not available from this repository. (Request a copy)


In this paper, we propose a novel information theoretic approach to obtain compact and discriminative dictionary of visual data. This approach squeezes discriminative information from dictionary for efficient representation using information bottleneck. The dictionary is optimized from the initial sparse dictionary which is learned from action data. In this, a constraint information optimization problem is formulated in which mutual information between initial and optimized dictionary is minimized while maximizing mutual information between optimized dictionary and class labels. We use an effective similarity measure, Jensen-Shannon divergence with adaptive weightages, for class distributions of each dictionary atom.These adaptive weightages are obtained based on the usage of dictionary atom among different classes. The resultant dictionary becomes discriminative and compact, while retaining maximum information with fewer atoms. Using simple reconstruction error, we test computational efficiency of the proposed method without compromising classification accuracy on popular benchmark datasets. It is further demonstrated how efficiently discriminative information is retained by comparing the classification performance of the dictionary before and after the removal of redundant dictionary atoms.

[error in script]
IITH Creators:
IITH CreatorsORCiD
Item Type: Article
Uncontrolled Keywords: Dictionary learning, Sparse representation, Information bottleneck, Mutual Information
Subjects: Computer science > Special computer methods
Divisions: Department of Computer Science & Engineering
Depositing User: Team Library
Date Deposited: 05 Jul 2017 06:33
Last Modified: 01 Sep 2017 09:02
Publisher URL:
OA policy:
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 3342 Statistics for this ePrint Item