Neuromorphic computing has emerged as a promising avenue towards building the next generation of intelligent computing systems. It has been proposed that memristive devices, which exhibit ...history-dependent conductivity modulation, could efficiently represent the synaptic weights in artificial neural networks. However, precise modulation of the device conductance over a wide dynamic range, necessary to maintain high network accuracy, is proving to be challenging. To address this, we present a multi-memristive synaptic architecture with an efficient global counter-based arbitration scheme. We focus on phase change memory devices, develop a comprehensive model and demonstrate via simulations the effectiveness of the concept for both spiking and non-spiking neural networks. Moreover, we present experimental results involving over a million phase change memory devices for unsupervised learning of temporal correlations using a spiking neural network. The work presents a significant step towards the realization of large-scale and energy-efficient neuromorphic computing systems.
Conventional computers based on the von Neumann architecture perform computation by repeatedly transferring data between their physically separated processing and memory units. As computation becomes ...increasingly data centric and the scalability limits in terms of performance and power are being reached, alternative computing paradigms with collocated computation and storage are actively being sought. A fascinating such approach is that of computational memory where the physics of nanoscale memory devices are used to perform certain computational tasks within the memory unit in a non-von Neumann manner. We present an experimental demonstration using one million phase change memory devices organized to perform a high-level computational primitive by exploiting the crystallization dynamics. Its result is imprinted in the conductance states of the memory devices. The results of using such a computational memory for processing real-world data sets show that this co-existence of computation and storage at the nanometer scale could enable ultra-dense, low-power, and massively-parallel computing systems.
This paper presents a dual-stage approach to nanopositioning in which the tradeoff between the scanner speed and range is addressed by combining a slow, large-range scanner with a short-range scanner ...optimized for high-speed, high-resolution positioning. We present the design, finite-element simulations, and experimental characterization of a fast custom-built short-range scanner. The short-range scanner is based on electromagnetic actuation to provide high linearity, has a clean, high-bandwidth dynamical response and is equipped with a low-noise magnetoresistance-based sensor. By using advanced noise-resilient feedback controllers, the dual-stage system allows large-range positioning with subnanometer closed-loop resolution over a wide bandwidth. Experimental results are presented in which the dual-stage scanner system is used for imaging in a custom-built atomic force microscope.
Stochastic phase-change neurons Tuma, Tomas; Pantazi, Angeliki; Le Gallo, Manuel ...
Nature nanotechnology,
08/2016, Letnik:
11, Številka:
8
Journal Article
Recenzirano
Artificial neuromorphic systems based on populations of spiking neurons are an indispensable tool in understanding the human brain and in constructing neuromimetic computational systems. To reach ...areal and power efficiencies comparable to those seen in biological systems, electroionics-based and phase-change-based memristive devices have been explored as nanoscale counterparts of synapses. However, progress on scalable realizations of neurons has so far been limited. Here, we show that chalcogenide-based phase-change materials can be used to create an artificial neuron in which the membrane potential is represented by the phase configuration of the nanoscale phase-change device. By exploiting the physics of reversible amorphous-to-crystal phase transitions, we show that the temporal integration of postsynaptic potentials can be achieved on a nanosecond timescale. Moreover, we show that this is inherently stochastic because of the melt-quench-induced reconfiguration of the atomic structure occurring when the neuron is reset. We demonstrate the use of these phase-change neurons, and their populations, in the detection of temporal correlations in parallel data streams and in sub-Nyquist representation of high-bandwidth signals.
As the conventional von Neumann-based computational architectures reach their scalability and performance limits, alternative computational frameworks inspired by biological neuronal networks hold ...promise to revolutionize the way we process information. Here, we present a bioinspired computational primitive that utilizes an artificial spiking neuron equipped with plastic synapses to detect temporal correlations in data streams in an unsupervised manner. We demonstrate that the internal states of the neuron and of the synapses can be efficiently stored in nanoscale phase-change memory devices and show computations with collocated storage in an experimental setting.
Nanopositioning is a key enabling technology for nanoscale science and engineering. Many nanopositioning systems employ feedback control to guarantee precise and repeatable positioning. However, ...achieving the desired performance with conventional feedback systems has remained a challenge. This paper analyzes a novel hybrid control architecture for high-speed nanopositioning, which is based on impulsive control. By impulsively changing the states of the feedback controller, performance objectives can be met that are beyond the limitations of linear feedback. We analyze the stability and performance of impulsive feedback control, and present experimental results in which impulsive control is used for precise motion control in a high-speed scanning-probe microscope.
Nanopositioning is a key enabling technology for nanoscale metrology and manipulation. This paper details experimental studies aimed at achieving high-bandwidth nanopositioning through a combination ...of scanner design with excellent dynamical behavior, novel high-bandwidth position sensing, and modern control techniques. Through a combination of high stiffness/rigidity of the flexures, a low carried mass, and uncomplicated mechanical connections, an X/Y scanner is designed which has the first resonant frequencies beyond 4kHz in both scan axes. For closed-loop operation of such fast scanners, there is a need for high-bandwidth, low-noise sensing schemes. A sensing concept based on magnetoresistance is presented that shows great potential towards providing low-noise position sensing over a very wide bandwidth. Atomic force microscopy imaging experiments of nanoscale structures are presented to illustrate the frame-per-second imaging capability of the nanopositioning system.
Hepatocellular carcinoma (HCC) represents one of the most common cancers worldwide. A considerable proportion of HCC is caused by cirrhosis related to metabolic dysfunction-associated steatohepatitis ...(MASH). Due to the increasing prevalence of metabolic syndrome, it is estimated that MASH-related HCC will become the most prevalent etiology of HCC. Currently, HCC screening is based on liver ultrasonography; however, the sensitivity of ultrasonography for early HCC stages in obese patients only reaches 23 %. To date, no studied biomarker shows sufficient efficacy for screening purposes. Nevertheless, the usage of spectroscopic methods offers a new perspective, as its potential use would provide cheap, fast analysis of samples such as blood plasma.
We employed a combination of conventional and chiroptical spectroscopic methods to study differences between the blood plasma of obese cirrhotic patients with and without HCC. We included 20 subjects with HCC and 17 without evidence of liver cancer, all of them with body mass index ≥ 30.
Sensitivities and specificities reached values as follows: 0.780 and 0.905 for infrared spectroscopy, 0.700 and 0.767 for Raman spectroscopy, 0.840 and 0.743 for electronic circular dichroism, and 0.805 and 0.923 for Raman optical activity. The final combined classification model based on all spectroscopic methods reached a sensitivity of 0.810 and a specificity of 0.857, with the highest area under the receiver operating characteristic curve among all models (0.961).
We suggest that this approach can be used effectively as a diagnostic tool in patients who are not examinable by liver ultrasonography.
NCT04221347