Dictionary Learning from Ambiguously Labeled Data

Yi-Chen Chen, Vishal M. Patel, Jaishanker K. Pillai, Rama Chellappa, P. J. Phillips; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2013, pp. 353-360

Abstract


We propose a novel dictionary-based learning method for ambiguously labeled multiclass classification, where each training sample has multiple labels and only one of them is the correct label. The dictionary learning problem is solved using an iterative alternating algorithm. At each iteration of the algorithm, two alternating steps are performed: a confidence update and a dictionary update. The confidence of each sample is defined as the probability distribution on its ambiguous labels. The dictionaries are updated using either soft (EM-based) or hard decision rules. Extensive evaluations on existing datasets demonstrate that the proposed method performs significantly better than state-of-the-art ambiguously labeled learning approaches.

Related Material


[pdf]
[bibtex]
@InProceedings{Chen_2013_CVPR,
author = {Chen, Yi-Chen and Patel, Vishal M. and Pillai, Jaishanker K. and Chellappa, Rama and Phillips, P. J.},
title = {Dictionary Learning from Ambiguously Labeled Data},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2013}
}