Abstract:Despite all the progress of semiconductor integrated circuit technology, the extreme complexity of the human cerebral cortex makes the hardware implementation of neuromorphic networks with a comparable number of devices exceptionally challenging. One of the most prospective candidates to provide comparable complexity, while operating much faster and with manageable power dissipation, are so-called CrossNets based on hybrid CMOS/memristor circuits. In these circuits, the usual CMOS stack is augmented with one or several crossbar layers, with adjustable two-terminal memristors at each crosspoint. Recently, there was a significant progress in improvement of technology of fabrication of such memristive crossbars and their integration with CMOS circuits, including first demonstrations of their vertical integration. Separately, there have been several demonstrations of discrete memristors as artificial synapses for neuromorphic networks. Very recently such experiments were extended to crossbar arrays of phase-change memristive devices. The adjustment of such devices, however, requires an additional transistor at each crosspoint, and hence the prospects of their scaling are less impressive than those of metal-oxide memristors, whose nonlinear I-V curves enable transistor-free operation. Here we report the first experimental implementation of a transistor-free metal-oxide memristor crossbar with device variability lowered sufficiently to demonstrate a successful operation of a simple integrated neural network, a single layer-perceptron. The network could be taught in situ using a coarse-grain variety of the delta-rule algorithm to perform the perfect classification of 3x3-pixel black/white images into 3 classes. We believe that this demonstration is an important step towards the implementation of much larger and more complex memristive neuromorphic networks.

Architectures for Nanoelectronic Implementation of Artificial Neural Networks

Architectures for Nanoelectronic Implementation of Artificial Neural Networks: New Results.

Architectures for Nanoelectronic Neural Networks: New Results.

Neuromorphic Architectures For Hybrid Nanoelectronic Circuits

Neuromorphic Architectures for Nanoelectronic Circuits

CMOL CrossNets: Possible Neuromorphic Nanoelectronic Circuits

Hybrid Dual-Complementary Metal-Oxide-semiconductor/memristor Synapse-Based Neural Network with Its Applications in Image Super-Resolution

3D Cmol Crossnet for Neuromorphic Network Applications

NEUTRAMS: Neural Network Transformation and Co-Design under Neuromorphic Hardware Constraints

Design and Implementation of a Flexible Neuromorphic Computing System for Affective Communication Via Memristive Circuits

Scalable Digital Neuromorphic Architecture for Large-Scale Biophysically Meaningful Neural Network With Multi-Compartment Neurons

End-to-End Implementation of Various Hybrid Neural Networks on a Cross-Paradigm Neuromorphic Chip

Training and Operation of an Integrated Neuromorphic Network Based on Metal-Oxide Memristors

A New Hybrid Neural System Interfacing Neurons and Silicon Hardware for Fast Signal Recognition

Molecular cellular networks: A non von Neumann architecture for molecular electronics

Hardware Realization of Building Blocks for Artificial Neural Networks

DiaNet: An Efficient Multi-Grained Re-configurable Neural Network in Silicon

A heterogeneous computing system with memristor-based neuromorphic accelerators

Accelerating Hybrid And Compact Neural Networks Targeting Perception And Control Domains With Coarse-Grained Dataflow Reconfiguration

Nanotube Devices Based Crossbar Architecture: Toward Neuromorphic Computing.

Low-voltage solution-processed artificial optoelectronic hybrid-integrated neuron based on 2D MXene for multi-task spiking neural network