判别式
计算机科学
人工智能
模式识别(心理学)
词典学习
代表(政治)
K-SVD公司
约束(计算机辅助设计)
规范(哲学)
稀疏逼近
机器学习
数学
几何学
政治
政治学
法学
作者
Shuhang Gu,Lei Zhang,Wangmeng Zuo,Xiangchu Feng
出处
期刊:Neural Information Processing Systems
日期:2014-12-08
卷期号:27: 793-801
被引量:262
摘要
Discriminative dictionary learning (DL) has been widely studied in various pattern classification problems. Most of the existing DL methods aim to learn a synthesis dictionary to represent the input signal while enforcing the representation coefficients and/or representation residual to be discriminative. However, the l0 or l1-norm sparsity constraint on the representation coefficients adopted in most DL methods makes the training and testing phases time consuming. We propose anew discriminative DL framework, namely projective dictionary pair learning (DPL), which learns a synthesis dictionary and an analysis dictionary jointly to achieve the goal of signal representation and discrimination. Compared with conventional DL methods, the proposed DPL method can not only greatly reduce the time complexity in the training and testing phases, but also lead to very competitive accuracies in a variety of visual classification tasks.
科研通智能强力驱动
Strongly Powered by AbleSci AI