Advancing neural computation: experimental validation and optimization of dendritic learning in feedforward tree networks DOI
Seyed‐Ali Sadegh‐Zadeh,

Pooya Hazegh

American Journal of Neurodegenerative Disease, Journal Year: 2024, Volume and Issue: 13(5), P. 49 - 69

Published: Jan. 1, 2024

This study aims to explore the capabilities of dendritic learning within feedforward tree networks (FFTN) in comparison traditional synaptic plasticity models, particularly context digit recognition tasks using MNIST dataset. We employed FFTNs with nonlinear segment amplification and Hebbian rules enhance computational efficiency. The dataset, consisting 70,000 images handwritten digits, was used for training testing. Key performance metrics, including accuracy, precision, recall, F1-score, were analysed. models significantly outperformed plasticity-based across all metrics. Specifically, framework achieved a test accuracy 91%, compared 88% demonstrating superior classification. Dendritic offers more powerful by closely mimicking biological neural processes, providing enhanced efficiency scalability. These findings have important implications advancing both artificial intelligence systems neuroscience.

Language: Английский

Self-organized and self-sustained ensemble activity patterns in simulation of mouse primary motor cortex DOI Creative Commons
Donald Doherty, Jae-Ho Jung, Salvador Durá-Bernal

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 14, 2025

The idea of self-organized signal processing in the cerebral cortex has become a focus research since Beggs and Plentz 1 reported avalanches local field potential recordings from organotypic cultures acute slices rat somatosensory cortex. How intrinsically organizes signals remains unknown. A current hypothesis was proposed by condensed matter physicists Bak, Tang, Wiesenfeld 2 when they conjectured that if neuronal avalanche activity followed inverse power law distributions, then brain may be set around phase transitions within signals. We asked we would observe an isolated slice our data driven detailed simulation mouse primary motor cortex? If did, with power-law distributions size duration what look like? Our results demonstrate brief unstructured stimulus (100ms, 57 μ current) to small subset neurons (about 181 more than 10,000) simulated self-sustained values similar those vivo vitro experiments. observed 4 cross-layer cross-neuron population patterns, 3 which displayed dominant rhythmic component. Avalanches were each composed one or patterns.

Language: Английский

Citations

0
Physiological features of parvalbumin-expressing GABAergic interneurons contributing to high-frequency oscillations in the cerebral cortex DOI Creative Commons
Katarina Miličević,

Brianna L. Barbeau,

Darko D. Lovic

et al.

Current Research in Neurobiology, Journal Year: 2023, Volume and Issue: 6, P. 100121 - 100121

Published: Dec. 16, 2023

Parvalbumin-expressing (PV+) inhibitory interneurons drive gamma oscillations (30-80 Hz), which underlie higher cognitive functions. In this review, we discuss two groups/aspects of fundamental properties PV+ interneurons. the first group (dubbed

Language: Английский

Citations

4
Generation and Characterization of Three Novel Mouse Mutant Strains Susceptible to Audiogenic Seizures DOI Creative Commons
Elena G. Varlamova,

Vera P. Kuldaeva,

Natalia Mitina

et al.

Cells, Journal Year: 2024, Volume and Issue: 13(21), P. 1747 - 1747

Published: Oct. 22, 2024

The mechanisms of epileptogenesis after brain injury, ischemic stroke, or tumors have been extensively studied. As a result, many effective antiseizure drugs developed. However, there are still patients who resistant to therapy. molecular and genetic bases regarding such drug-resistant seizures poorly elucidated. In cases, heavy instigated by development malformations often caused gene mutations. Such can be demonstrated in mouse models generating mutant strains. One the most potent mutagens is ENU (N-ethyl-N-nitrosourea). present study, we describe three novel strains generated ENU-directed mutagenesis. Two these very strong epileptic phenotype triggered audiogenic stimuli (G9-1 S5-1 strains). third strain characterized behavioral disorders hyperexcitation neuronal networks. We identified changes expression those genes encoding neurotransmission proteins cerebral cortexes mice. It turned out that G9-1 strongest disruptions plasma membrane channels, excitatory glutamate receptors, protein kinases. On other hand, number GABAergic neurons was also affected mutation. All lines increased anxiety, excitability, suppressed motor orientational-exploratory activities. with an phenotype-G9-1 S5-1ave reduced learning ability, A9-2 mice line retains high ability.

Language: Английский

Citations

0
Advancing neural computation: experimental validation and optimization of dendritic learning in feedforward tree networks DOI
Seyed‐Ali Sadegh‐Zadeh,

Pooya Hazegh

American Journal of Neurodegenerative Disease, Journal Year: 2024, Volume and Issue: 13(5), P. 49 - 69

Published: Jan. 1, 2024

This study aims to explore the capabilities of dendritic learning within feedforward tree networks (FFTN) in comparison traditional synaptic plasticity models, particularly context digit recognition tasks using MNIST dataset. We employed FFTNs with nonlinear segment amplification and Hebbian rules enhance computational efficiency. The dataset, consisting 70,000 images handwritten digits, was used for training testing. Key performance metrics, including accuracy, precision, recall, F1-score, were analysed. models significantly outperformed plasticity-based across all metrics. Specifically, framework achieved a test accuracy 91%, compared 88% demonstrating superior classification. Dendritic offers more powerful by closely mimicking biological neural processes, providing enhanced efficiency scalability. These findings have important implications advancing both artificial intelligence systems neuroscience.

Language: Английский

Citations

0