Journal of Applied Physics · 2018 · 167 citations · 40 references
EngineeringComputer EngineeringElectrical EngineeringElectronic MaterialsMemorySemiconductor DevicesAmorphous Electronic MaterialsDevice ModelingMemory DevicesMemory ModelingPhase TransitionsSpiking Neural NetworksConductanceIntra-device VariabilityNeuromorphic DevicesNeuromorphic HardwareNon-volatile MemoryNonvolatile Memory DevicesPhase-change MaterialsPhase-change Memory
Phase‑change memory (PCM) is a non‑volatile technology that uses rapid amorphous‑to‑crystalline transitions to controllably adjust conductance, making it attractive for synaptic elements in neuromorphic systems, though its progressive crystallization and conductance drift introduce variability. The authors aim to develop a statistical model that captures both the cumulative conductance evolution and temporal drift inherent to PCM devices. The model is derived from extensive characterization of 10,000 PCM devices. Using the model, the authors simulate supervised training of both spiking and non‑spiking artificial neuronal networks.
Phase-change memory (PCM) is an emerging non-volatile memory technology that is based on the reversible and rapid phase transition between the amorphous and crystalline phases of certain phase-change materials. The ability to alter the conductance levels in a controllable way makes PCM devices particularly well-suited for synaptic realizations in neuromorphic computing. A key attribute that enables this application is the progressive crystallization of the phase-change material and subsequent increase in device conductance by the successive application of appropriate electrical pulses. There is significant inter- and intra-device randomness associated with this cumulative conductance evolution, and it is essential to develop a statistical model to capture this. PCM also exhibits a temporal evolution of the conductance values (drift), which could also influence applications in neuromorphic computing. In this paper, we have developed a statistical model that describes both the cumulative conductance evolution and conductance drift. This model is based on extensive characterization work on 10 000 memory devices. Finally, the model is used to simulate the supervised training of both spiking and non-spiking artificial neuronal networks.
40
Efficient Processing of Deep Neural Networks: A Tutorial and Survey
Vivienne Sze, Yu‐Hsin Chen, Tien-Ju Yang et al. · Proceedings of the IEEE · 2017 · 4K citations
Equivalent-accuracy accelerated neural-network training using analogue memory
Stefano Ambrogio, Pritish Narayanan, Hsinyu Tsai et al. · Nature · 2018 · 1.1K citations · Full text
Stochastic phase-change neurons
Tomáš Tůma, Angeliki Pantazi, Manuel Le Gallo et al. · Nature Nanotechnology · 2016 · 1.1K citations