Cargando…
Biologically plausible learning in recurrent neural networks reproduces neural dynamics observed during cognitive tasks
Neural activity during cognitive tasks exhibits complex dynamics that flexibly encode task-relevant variables. Chaotic recurrent networks, which spontaneously generate rich dynamics, have been proposed as a model of cortical computation during cognitive tasks. However, existing methods for training...
Autor principal: | Miconi, Thomas |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
eLife Sciences Publications, Ltd
2017
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5398889/ https://www.ncbi.nlm.nih.gov/pubmed/28230528 http://dx.doi.org/10.7554/eLife.20899 |
Ejemplares similares
-
Training biologically plausible recurrent neural networks on cognitive tasks with long-term dependencies
por: Soo, Wayne W.M., et al.
Publicado: (2023) -
Recurrent neural networks enable design of multifunctional synthetic human gut microbiome dynamics
por: Baranwal, Mayank, et al.
Publicado: (2022) -
Stable task information from an unstable neural population
por: Rule, Michael E, et al.
Publicado: (2020) -
Model discovery to link neural activity to behavioral tasks
por: Costabile, Jamie D, et al.
Publicado: (2023) -
Learning recurrent dynamics in spiking networks
por: Kim, Christopher M, et al.
Publicado: (2018)