Title :
Global convergence of independent component analysis based on semidefinite programming relaxation
Author :
Akaho, Shotaro ; Fujiki, Jun
Author_Institution :
Nat. Inst. of Adv. Ind. Sci. & Technol., Tsukuba, Japan
Abstract :
In the independent component analysis, polynomial functions of higher order statistics are often used as cost functions. However, such cost functions usually have many local minima, hence gradient-type and fixed-point-type algorithms tend to be trapped into a nonglobal local minimum. Recently, the polynomial optimization method that guarantees global convergence has been developed, where the optimization problem is relaxed as a semidefinite programming problem. In this paper, we apply the polynomial optimization method to the independent component analysis, and show the global convergence property. From some empirical studies, we further give a conjecture that the algorithm has polynomial time computational complexity.
Keywords :
computational complexity; higher order statistics; independent component analysis; mathematical programming; fixed-point-type algorithms; global convergence; gradient-type algorithms; higher order statistics; independent component analysis; nonglobal local minimum; polynomial functions; polynomial optimization method; polynomial time computational complexity; semidefinite programming relaxation; Indexes; Optimization; Polynomials; global convergence; independent component analysis; polynomial optimization; semidefinite programming;
Conference_Titel :
Acoustics, Speech and Signal Processing (ICASSP), 2011 IEEE International Conference on
Conference_Location :
Prague
Print_ISBN :
978-1-4577-0538-0
Electronic_ISBN :
1520-6149
DOI :
10.1109/ICASSP.2011.5947295