単一の錐体細胞モデルを用いた発話認識
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
How rich information is encoded and utilized for brain computation within a single neuron dynamically? We propose a novel method for constructing a decoder using the framework of physical reservoir computing. In this study, we performed a speech recognition task using a multi-compartmental model of a hippocampal CA1 pyramidal cell as a reservoir layer. The audio waveform was encoded in the activity of synapses randomly distributed along the dendrite. Results showed that intracellular calcium dynamics driven by synaptic activity can be harnessed for label prediction.
Fields of Study & Subjects
Neural Networks and Reservoir Computing
Neural dynamics and brain function
Advanced Memory and Neural Computing
Publication details
- Journal / source
- Institutional Repositories DataBase (IRDB)
- Volume
- 125
- Issue
- 344
- Pages
- 151–156
- Publication date
- 21 January 2026
- Language
- JA