Effect of conductance linearity of Ag-chalcogenide CBRAM synaptic devices on the pattern recognition accuracy of an analog neural training accelerator.
Priyanka ApsangiHugh J. BarnabyMichael N. KozickiYago Gonzalez-VeloJennifer TaggartPublished in: Neuromorph. Comput. Eng. (2022)
Keyphrases
- recognition accuracy
- recognition rate
- face recognition
- hebbian learning
- single neuron
- associative memory
- learning rules
- neuron model
- pattern matching
- neural network
- training set
- mobile devices
- spiking neurons
- spiking neural networks
- improve the recognition rate
- recurrent networks
- input patterns
- discriminating power
- network architecture
- detection rate
- principal component analysis
- training data
- machine learning