Model and experiments.
<p>A. Structure of the dataset: binary vectors are organized along a temporal scale. Some of the vectors repeat after time steps, others are randomly generated. Repeated vectors are considered familiar and are labeled as 1, non-repeated vectors are non-familiar and labeled as 0. <b>B<...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | , |
| Published: |
2025
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | <p>A. Structure of the dataset: binary vectors are organized along a temporal scale. Some of the vectors repeat after time steps, others are randomly generated. Repeated vectors are considered familiar and are labeled as 1, non-repeated vectors are non-familiar and labeled as 0. <b>B</b>. Three degrees of input sparseness are used in the experiment: 0.6, 0.8, 0.9. <b>C</b>. Inputs with correlation level modeled as template similarity, i.e., the fraction of externally stimulated neurons matching those in the same template activation pattern: 0.0 (main experiments), 0.2, 0.4, 0.6, 0.8 (see Methods Correlated inputs). <b>D</b>. The model architecture: each Izhikevich neuron [<a href="http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1013304#pcbi.1013304.ref031" target="_blank">31</a>] has a one-to-one connection to the spiking input. Izhikevich neurons are laterally connected to one another. Lateral connections undergo unidirectional STDP or anti-STDP.</p> |
|---|