Papers accepted for ISCAS 2018 in Florence, Italy

The Neuromorphic Behaving Systems Group is proud that two papers co-authored by Elisabetta Chicca are accepted to be presented at the IEEE International Symposium on Circuits and Systems 2018 in Florence, Italy, May 27-30, 2018.

Both papers deal with the topic of RRAM implementations for learning tasks using spike-timing dependent plasticity (STDP) learning rules.

Hereby the paper "Resistive switching synapses for unsupervised learning in feed-forward and recurrent neural networks" addresses the development of RRAM synapses for unsupervised learning. The application of a one-transistor/one-resistor structure for long term potentiation and depression with standard memory-grade RRAM devices is implemented alongside a hardware demonstration of unsupervised learning. 

The paper "Brain-inspired recurrent neural network with plastic RRAM synapses" deals with the design and simulation of a Hopfield-type recurrent neural network with HfO2 RRAM devices which are capable of learning via STDP. The implementation of a 64-neuron recurrent neural network (RNN) supports the application of RRAM-based RNNs to the modelling of cognitive computing functionalities.