Search results
- 1.0404257 - UIVT-O 20010080 CZ eng V - Research Report
Šíma, Jiří
Training a Sigmoid Neuron is Hard.
Prague: ICS AS CR, 2001. 12 s. Technical Report, V-835.
R&D Projects: GA ČR GA201/00/1489; GA AV ČR IAB2030007
Institutional research plan: AV0Z1030915
Keywords : loading problem * learning complexity * NP-hardness * sigmoid neuron * back-propagation * constructive learning
Subject RIV: BA - General Mathematics
Permanent Link: http://hdl.handle.net/11104/0124520File Download Size Commentary Version Access v835-01.pdf 18 214 KB Other open-access - 2.0404138 - UIVT-O 20000236 CZ eng V - Research Report
Hakl, František - Kalous, R.
Vzorová složitost teorie učení a PAC modelu.
[Sample's complexity and the PAC Learning Model.]
Prague: ICS AS CR, 2000. 22 s. Technical Report, V-825.
R&D Projects: GA AV ČR IAB2030007
Institutional research plan: AV0Z1030915
Permanent Link: http://hdl.handle.net/11104/0124407File Download Size Commentary Version Access v825-00.pdf 18 376 KB Other open-access - 3.0404137 - UIVT-O 20000235 CZ eng V - Research Report
Hakl, František
Možnosti aproximace funkcí neuronovými sítěmi.
[Approximation Properties of Neural Networks.]
Prague: ICS AS CR, 2000. 23 s. Technical Report, V-824.
R&D Projects: GA AV ČR IAB2030007
Institutional research plan: AV0Z1030915
Keywords : approximation * Kolmogorov theorem * neural networks
Permanent Link: http://hdl.handle.net/11104/0124406File Download Size Commentary Version Access v824-00.pdf 18 412.2 KB Other open-access - 4.0404072 - UIVT-O 20000197 CZ eng V - Research Report
Šíma, Jiří
The Computational Theory of Neural Networks.
Prague: ICS AS CR, 2000. 31 s. Technical Report, V-823.
R&D Projects: GA ČR GA201/98/0717; GA AV ČR IAB2030007
Institutional research plan: AV0Z1030915
Keywords : neural networks * perceptrons * RBF networks * winner-take-all networks * spiking neurons * computational power
Permanent Link: http://hdl.handle.net/11104/0124344File Download Size Commentary Version Access v823-00.pdf 18 342.6 KB Other open-access - 5.0404071 - UIVT-O 20000196 CZ eng V - Research Report
Šíma, Jiří
On the Computational Power of Continuous-Time Symmetric Hopfield Nets.
Prague: ICS AS CR, 2000. 37 s. Technical Report, V-815.
R&D Projects: GA ČR GA201/98/0717; GA AV ČR IAB2030007
Institutional research plan: AV0Z1030915
Keywords : neural networks * continuous-time Hopfield net * Lyapunov function * analog computation * computational power
Permanent Link: http://hdl.handle.net/11104/0124343File Download Size Commentary Version Access v815-00.pdf 18 375.6 KB Other open-access