TY - GEN
T1 - Radiation Effect on Learning Behavior in Memristor-Based Neuromorphic Circuit
AU - Dahl, Sumedha Gandharava
AU - Ivans, Rober C.
AU - Cantley, Kurtis D.
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/8
Y1 - 2019/8
N2 - In this work, a memristor-based spiking neural network with a many-to-one feed-forward topology is designed for spatio-temporal pattern learning (25-pixel character 'B'). A TiO2 non-linear drift behavioral memristor model is used for simulation of a neuromorphic circuit with 25 pre- and 1 post-synaptic neuron. The memristor model mimics synaptic behavior and is modified to include effects of radiation interactions on memristor. Simulated irradiation of the memristors during the learning process alters their state, and the effect on circuit learning behavior is observed. The network uses spike-timing-dependent plasticity (STDP) learning implemented using biphasic neuron action potentials. It is found that although radiation changes the synaptic weights, the network is able to recover and relearn the pattern after radiation ceases. Recovery time is proportional to flux, intensity, and duration of the radiation. If the network is exposed to radiation events of low intensity and flux even for an indefinite period (as observed in space), it may be able to retain its pattern recognition capabilities.
AB - In this work, a memristor-based spiking neural network with a many-to-one feed-forward topology is designed for spatio-temporal pattern learning (25-pixel character 'B'). A TiO2 non-linear drift behavioral memristor model is used for simulation of a neuromorphic circuit with 25 pre- and 1 post-synaptic neuron. The memristor model mimics synaptic behavior and is modified to include effects of radiation interactions on memristor. Simulated irradiation of the memristors during the learning process alters their state, and the effect on circuit learning behavior is observed. The network uses spike-timing-dependent plasticity (STDP) learning implemented using biphasic neuron action potentials. It is found that although radiation changes the synaptic weights, the network is able to recover and relearn the pattern after radiation ceases. Recovery time is proportional to flux, intensity, and duration of the radiation. If the network is exposed to radiation events of low intensity and flux even for an indefinite period (as observed in space), it may be able to retain its pattern recognition capabilities.
KW - leaky integrate-and-fire (LIF) neuron
KW - Neuromorphic circuit
KW - nonlinear memristor model
KW - spatio-temporal pattern
KW - spike-timing-dependence plasticity (STDP)
KW - state-altering radiation
UR - http://www.scopus.com/inward/record.url?scp=85074995255&partnerID=8YFLogxK
U2 - 10.1109/MWSCAS.2019.8885288
DO - 10.1109/MWSCAS.2019.8885288
M3 - Conference contribution
AN - SCOPUS:85074995255
T3 - Midwest Symposium on Circuits and Systems
SP - 53
EP - 56
BT - 2019 IEEE 62nd International Midwest Symposium on Circuits and Systems, MWSCAS 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 62nd IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2019
Y2 - 4 August 2019 through 7 August 2019
ER -