Abstract
Brain-inspired computing architectures attempt to mimic the computations performed in the neurons and the synapses in the human brain in order to achieve its efficiency in learning and cognitive tasks. In this work, we demonstrate the mapping of the probabilistic spiking nature of pyramidal neurons in the cortex to the stochastic switching behavior of a Magnetic Tunnel Junction in presence of thermal noise. We present results to illustrate the efficiency of neuromorphic systems based on such probabilistic neurons for pattern recognition tasks in presence of lateral inhibition and homeostasis. Such stochastic MTJ neurons can also potentially provide a direct mapping to the probabilistic computing elements in Belief Networks for performing regenerative tasks.
Highlights
Brain-inspired computing architectures attempt to mimic the computations performed in the neurons and the synapses in the human brain in order to achieve its efficiency in learning and cognitive tasks
We demonstrate the mapping of the probabilistic spiking nature of pyramidal neurons in the cortex to the stochastic switching behavior of a Magnetic Tunnel Junction in presence of thermal noise
To the best of our knowledge, this is the first demonstration of mapping the stochastic leaky-integrate switching behavior of Magnetic Tunnel Junction (MTJ) in presence of thermal noise to a probabilistic spiking neuron
Summary
Past research on hardware implementation of spiking neurons have mainly focused on deterministic neural models, like the Hodgkin-Huxley[1] and Leaky-Integrate-Fire[1] models Emulation of such neural characteristics require area-expensive CMOS implementations involving more than 20 transistors[2,3] and a direct mapping of spiking neuronal characteristics to a single nanoelectronic device is still missing. Such deterministic neuron models have little correspondence to the probabilistic firing nature of biological neurons and are unable to account for the fact that neural computation in the brain is significantly prone to noise arising from the synapses, dendrites or the neuron itself[4,5]. We demonstrate a nano-magnetic device that can mimic such cortical “stochastic” spiking neurons
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call