大会名称 |
---|
2016年 ソサイエティ大会 |
大会コ-ド |
2016S |
開催年 |
2016 |
発行日 |
2016-09-06 |
セッション番号 |
BS-5 |
セッション名 |
Network and Service Design, Control and Management |
講演日 |
2016/9/22 |
講演場所(会議室等) |
工学部 M棟 M151 |
講演番号 |
BS-5-26 |
タイトル |
IMAGE CLASSIFIER LEARNING FROM NOISY LABELS VIA GENERALIZED GRAPH SMOOTHNESS PRIORS |
著者名 |
○Yu Mao, Gene Cheung, Chia-wen Lin, Yusheng Ji, |
キーワード |
classification |
抄録 |
When collecting samples via crowd-sourcing for semi-supervised learning, often labels that designate events of interest are assigned unreliably, resulting in label noise. In this paper, we propose a robust method for graph-based image classifier learning given noisy labels, leveraging on recent advances in graph signal processing. In particular, we formulate a graph-signal restoration problem, where the objective includes a fidelity term to minimize the l0-norm between the observed labels and a reconstructed graph-signal, and generalized graph smoothness priors, where we assume that the reconstructed signal and its gradient are both smooth with respect to a graph. Simulation results show that our proposed algorithm can outperform both regular SVM and robust label noise learning approaches in the literature noticeably. |
本文pdf |
PDF download
|