Naslov
Unsupervised Learning with Stochastic Gradient

Autori
Szu, Harold ; Kopriva, Ivica

Izvornik
Neurocomputing (0925-2312) 68 (2005); 130-160

Vrsta, podvrsta i kategorija rada
Radovi u časopisima, članak, znanstveni

Ključne riječi
stochastic optimization; Cauchy annealing; blind source separation; Helmholtz free energy

Sažetak
A stochastic gradient is formulated based on deterministic gradient augmented with Cauchy simulated annealing capable to reach a global minimum with a convergence speed significantly faster then when simulated annealing is used alone. In order to solve space-time variant inverse problems known as blind source separation, a novel Helmholtz free energy contrast function, , with imposed thermodynamics constraint at a constant temperature T0 was introduced generalizing the Shannon maximum entropy S of the closed systems to the open systems having non-zero input-output energy exchange E. Here only the input data vector was known while source vector and mixing matrix were unknown. A stochastic gradient was successfully applied to solve inverse space-variant imaging problems on a concurrent pixel-by-pixel basis with the unknown mixing matrix (imaging point spread function) varying from pixel to pixel.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika

POVEZANOST RADA