The rapid growth of brain-inspired computing coupled with the inefficiencies in the CMOS implementations of neuromrphic systems has led to intense exploration of efficient hardware implementations of the functional units of the brain, namely, neurons and synapses. However, efforts have largely been invested in implementations in the electrical domain with potential limitations of switching speed, packing density of large integrated systems and interconnect losses. As an alternative, neuromorphic engineering in the photonic domain has recently gained attention. In this work, we propose a purely photonic operation of an Integrate-and-Fire Spiking neuron, based on the phase change dynamics of Ge2Sb2Te5 (GST) embedded on top of a microring resonator, which alleviates the energy constraints of PCMs in electrical domain. We also show that such a neuron can be potentially integrated with on-chip synapses into an all-Photonic Spiking Neural network inferencing framework which promises to be ultrafast and can potentially offer a large operating bandwidth.

Spiking Neural Networks (SNNs) offer an event-driven and more biologically realistic alternative to standard Artificial Neural Networks based on analog information processing. This can potentially enable energy-efficient hardware implementations of neuromorphic systems which emulate the functional units of the brain, namely, neurons and synapses. Recent demonstrations of ultra-fast photonic computing devices based on phase-change materials (PCMs) show promise of addressing limitations of electrically driven neuromorphic systems. However, scaling these standalone computing devices to a parallel in-memory computing primitive is a challenge. In this work, we utilize the optical properties of the PCM, Ge2Sb2Te5 (GST), to propose a Photonic Spiking Neural Network computing primitive, comprising of a non-volatile synaptic array integrated seamlessly with previously explored 'integrate-and-fire' neurons. The proposed design realizes an 'in-memory' computing platform that leverages the inherent parallelism of wavelength-division-multiplexing (WDM). We show that the proposed computing platform can be used to emulate a SNN inferencing engine for image classification tasks. The proposed design not only bridges the gap between isolated computing devices and parallel large-scale implementation, but also paves the way for ultra-fast computing and localized on-chip learning.

No abstract

Functional interfaces between electronics and biological matter are essential to diverse fields including health sciences and bio-engineering. Here, we report the discovery of spontaneous (no external energy input) hydrogen transfer from biological glucose reactions into SmNiO 3 , an archetypal perovskite quantum material. The enzymatic oxidation of glucose is monitored down to ~5 × 10 −16 M concentration via hydrogen transfer to the nickelate lattice. The hydrogen atoms donate electrons to the Ni d orbital and induce electron localization through strong electron correlations. By enzyme specific modification, spontaneous transfer of hydrogen from the neurotransmitter dopamine can be monitored in physiological media. We then directly interface an acute mouse brain slice onto the nickelate devices and demonstrate measurement of neurotransmitter release upon electrical stimulation of the striatum region. These results open up avenues for use of emergent physics present in quantum materials in trace detection and conveyance of bio-matter, bio-chemical sciences, and brain-machine interfaces.

No abstract

In-memory computing (IMC) is a promising approach for energy cost reduction due to data movement between memory and processor for running data-intensive deep learning applications on the computing systems. Together with Binary Neural Network (BNN), IMC provides a viable solution for running deep neural networks at the edge devices with stringent memory and energy constraints. In this paper, we propose a novel 10T bit-cell with a back-end-of-line (BEOL) metal-oxide-metal (MOM) capacitor laid on pitch for in-memory computing. Our IMC bit-cell, when arranged in a memory array, performs binary convolution (XNOR followed by Bit-count operations) and binary activation generation operations. We show, when binary layers of BNN are mapped into our IMC arrays for MNIST digit classification, 98.75% accuracy with energy efficiency of 2193 TOPS/W and throughput of 22857 GOPS can be obtained. We determine the memory array size considering the word-line and bit-line nonidealities and show how these impact classification accuracy. We analyze the impact of process variations on classification accuracy and show how word-line pulse tunability provided by our design can be used to improve the robustness of classification under process variations.

No abstract

The novel electronic properties of graphene nanoribbons (GNRs) including purely two-dimensional structure along with its tunable bandgap have led to intense research into possible applications of this material in nanoscale devices. However, as yet, dimensions of its possibilities in practical device levels have remained inconsistent. In this paper we propose a model for GNR-FET that is made from only Armchair GNRs. Our complete NEGF-based simulation reveals its potential for fast digital electronics with On/Off ratio up to 10 3 , transconductance of 8.5 10 3 nS/nm which lead to a analog operational frequency up to 3.3THz. The effects of Stone-Wales defects and Edge Roughness in GNRs have been analysed here that shows switching and frequency performance degradation of such GNR-FETs.