Effects of memristive synapse radiation interactions on learning in spiking neural networks
| Autor: | Kurtis D. Cantley, Robert C. Ivans, Sumedha Gandharava Dahl |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
Technology
Computer science General Chemical Engineering Science Stability (learning theory) Spatio-temporal pattern learning General Physics and Astronomy 02 engineering and technology Radiation Synapse Modeling and simulation 03 medical and health sciences Synaptic weight 0302 clinical medicine Leaky integrate-and-fire (LIF) neuron Learning rule General Materials Science General Environmental Science Spiking neural network Spike-timing-dependent plasticity (STDP) General Engineering 021001 nanoscience & nanotechnology Neuromorphic circuits General Earth and Planetary Sciences 0210 nano-technology Biological system 030217 neurology & neurosurgery Learning behavior Non-linear memristor model |
| Zdroj: | SN Applied Sciences, Vol 3, Iss 5, Pp 1-16 (2021) |
| ISSN: | 2523-3971 2523-3963 |
| Popis: | This study uses advanced modeling and simulation to explore the effects of external events such as radiation interactions on the synaptic devices in an electronic spiking neural network. Specifically, the networks are trained using the spike-timing-dependent plasticity (STDP) learning rule to recognize spatio-temporal patterns (STPs) representing 25 and 100-pixel characters. Memristive synapses based on a TiO2 non-linear drift model designed in Verilog-A are utilized, with STDP learning behavior achieved through bi-phasic pre- and post-synaptic action potentials. The models are modified to include experimentally observed state-altering and ionizing radiation effects on the device. It is found that radiation interactions tend to make the connection between afferents stronger by increasing the conductance of synapses overall, subsequently distorting the STDP learning curve. In the absence of consistent STPs, these effects accumulate over time and make the synaptic weight evolutions unstable. With STPs at lower flux intensities, the network can recover and relearn with constant training. However, higher flux can overwhelm the leaky integrate-and-fire post-synaptic neuron circuits and reduce stability of the network. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink