Nota de FísicaPublicado
Optimal Hebbian Learning Rules and the Role of Assymetry
D. A. Stariolo; C. Tsallis
1992
Dados da publicação
- Autores
- D. A. Stariolo; C. Tsallis
- Tipo
- Nota de Física
- Status
- Publicado
- Data de publicação
- 1992
- Descrição
- We study the storage properties associated with generalized Hebbian learning rules which present four free parameters that allow for asymmetry. We also introduce two extra parameters in the post-synaptic potentials in order to further improve the critical capacity. Using signal-to-noise analysis, as well as computer simulations on an analog network, we discuss the performance of the rules for arbitrarily biased patterns and find that the critical storage capacity <math>\alpha_c\,\!</math> becomes maximal for a particular symmetric rule ( <math>\alpha_c\,\!</math> diverges in the sparce coding limit). Departures from symmetry decrease <math>\alpha_c\,\!</math> but can increase the robustness of the model.
- Ano
- 1992
- Número
- CBPF-NF-038/92
- Vínculo com tese
- Não
- Vínculo com bolsa
- Não