Details
![Melika Payvand Headshot](https://confcats-catavault.s3.amazonaws.com/CATAVault/ieeecass/master/files/styles/cc_user_photo/s3/user-pictures/26041.jpg?h=fbf7a813&itok=4HwPNflK)
- Affiliation
-
AffiliationUniversity of Zurich/ ETH Zurich
- Country
Continuous learning using mixed-signal hybrid memristive-CMOS neuromorphic can resolve open challenges related to noise and variability which will enable their employment for IoT devices. To overcome the problems of variability and limited resolution of ReRAM devices storing the synaptic weights, we propose to only use their high conductive state and control their desired conductance by modulating their programming compliance current. We describe the spike-based learning CMOS circuits that are used to modulate the synaptic weights and demonstrate the relationship between the synaptic weight, the device conductance, and the compliance current used to set its weight, with experimental measurements from a 4kb array of HfO2-based devices. To validate the approach and the circuits presented, we present circuit simulation results for a standard CMOS 180nm process and system-level behavioral simulations for classifying hand-written digits from the MNIST data-set with classification accuracy of 92.68% on the test set.