Neuromorphic features of probabilistic neural networks

0090278 - ÚTIA 2008 RIV CZ eng J - Journal Article
Grim, Jiří
Neuromorphic features of probabilistic neural networks.
[Neuromorfní vlastnosti pravděpodobnostních neuronových sítí.]
Kybernetika. Roč. 43, č. 5 (2007), s. 697-712. ISSN 0023-5954
R&D Projects: GA ČR GA102/07/1594; GA MŠMT 1M0572
EU Projects: European Commission(XE) 507752 - MUSCLE
Grant - others:GA MŠk(CZ) 2C06019
Institutional research plan: CEZ:AV0Z10750506
Keywords : probabilistic neural networks * distribution mixtures * sequential EM algorithm * pattern recognition
Subject RIV: IN - Informatics, Computer Science
Impact factor: 0.552, year: 2007

We summarize the main results on probabilistic neural networks recently published in a series of papers. Considering the framework of statistical pattern recognition we assume approximation of class-conditional distributions by finite mixtures of product components. The probabilistic neurons correspond to mixture components and can be interpreted in neurophysiological terms. In this way we can find possible theoretical background of the functional properties of neurons. For example, the general formula for synaptical weights provides a statistical justification of the well known Hebbian principle of learning. Similarly, the mean effect of lateral inhibition can be expressed by means of a formula proposed by Perez as a measure of dependence tightness of involved variables.

Souhrnná práce o pravděpodobnostních neuronových sítích, které nabízejí alternativní řešení problému výběru příznaků (podprostorový přístup) a jsou široce použitelné pro řešení mnohorozměrných úloh klasifikace s omezenými datovými soubory.
Permanent Link: http://hdl.handle.net/11104/0151218