Title :
Optimal error for Orthogonal Matching Pursuit for μ-coherent dictionaries
Author :
Jingfan Long ; Xiujie Wei ; Peixin Ye
Author_Institution :
Beijing Inf. Sci. & Technol. Univ., Beijing, China
Abstract :
In this paper, we investigate the efficiency of some kind of Greedy Algorithms with respect to dictionaries from Hilbert spaces. We establish ideal Lebesgue-type inequality for Orthogonal Matching Pursuit which is also known as the Orthogonal Greedy Algorithm for μ-coherent dictionaries. We show that the Orthogonal Matching Pursuit provides an almost optimal approximation on the first [1/18μ] steps.
Keywords :
Hilbert spaces; approximation theory; error analysis; greedy algorithms; learning (artificial intelligence); μ-coherent dictionaries; Hilbert spaces; error analysis; ideal Lebesgue-type inequality; optimal approximation; optimal error; orthogonal greedy algorithm; orthogonal matching pursuit; Approximation methods; Coherence; Dictionaries; Educational institutions; Frequency modulation; Greedy algorithms; Matching pursuit algorithms;
Conference_Titel :
Natural Computation (ICNC), 2013 Ninth International Conference on
Conference_Location :
Shenyang
DOI :
10.1109/ICNC.2013.6818200
