A Parasitic Resistance-Adapted Programming Scheme for Memristor Crossbar-Based Neuromorphic Computing Systems

Truong, S.

doi:10.3390/ma12244097

Cited by 5 publications

(7 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Truong proposes a method to correct the line resistances when writing the desired values of conductance in DNN for feedforward applications [19]. Circuit simulations of a 64 × 16 single layer perceptron for the recognition of 26 characters (8 × 8 grayscale pixels) support significant improvements of network recognition when line resistance increases above 1.5Ω.…”

Section: Synopsismentioning

confidence: 98%

Memristors for Neuromorphic Circuits and Artificial Intelligence Applications

Miranda

Suñé

2020

Materials

View full text Add to dashboard Cite

Artificial Intelligence has found many applications in the last decade due to increased computing power. Artificial Neural Networks are inspired in the brain structure and consist in the interconnection of artificial neurons through artificial synapses in the so-called Deep Neural Networks (DNNs). Training these systems requires huge amounts of data and, after the network is trained, it can recognize unforeseen data and provide useful information. As far as the training is concerned, we can distinguish between supervised and unsupervised learning. The former requires labelled data and is based on the iterative minimization of the output error using the stochastic gradient descent method followed by the recalculation of the strength of the synaptic connections (weights) with the backpropagation algorithm. On the other hand, unsupervised learning does not require data labeling and it is not based on explicit output error minimization. Conventional ANNs can function with supervised learning algorithms (perceptrons, multi-layer perceptrons, convolutional networks, etc.) but also with unsupervised learning rules (Kohonen networks, self-organizing maps, etc.). Besides, another type of neural networks are the so-called Spiking Neural Networks (SNNs) in which learning takes place through the superposition of voltage spikes launched by the neurons. Their behavior is much closer to the brain functioning mechanisms they can be used with supervised and unsupervised learning rules. Since learning and inference is based on short voltage spikes, energy efficiency improves substantially. Up to this moment, all these ANNs (spiking and conventional) have been implemented as software tools running on conventional computing units based on the von Neumann architecture. However, this approach reaches important limits due to the required computing power, physical size and energy consumption. This is particularly true for applications at the edge of the internet. Thus, there is an increasing interest in developing AI tools directly implemented in hardware for this type of applications. The first hardware demonstrations have been based on Complementary Metal-Oxide-Semiconductor (CMOS) circuits and specific communication protocols. However, to further increase training speed andenergy efficiency while reducing the system size, the combination of CMOS neuron circuits with memristor synapses is now being explored. It has also been pointed out that the short time non-volatility of some memristors may even allow fabricating purely memristive ANNs. The memristor is a new device (first demonstrated in solid-state in 2008) which behaves as a resistor with memory and which has been shown to have potentiation and depression properties similar to those of biological synapses. In this Special Issue, we explore the state of the art of neuromorphic circuits implementing neural networks with memristors for AI applications.

show abstract

Section: Synopsismentioning

confidence: 98%

Memristors for Neuromorphic Circuits and Artificial Intelligence Applications

Miranda

Suñé

2020

Materials

View full text Add to dashboard Cite

show abstract

“…[ 285 ] Parasitic resistance mainly refers to wire resistance along the row and column lines in memristor crossbar arrays. [ 285 ] During the read operation, parasitic conductive paths in unselected cells will degrade the output signal and result in massive sneak path leakage which will further increase the possibility of misreading during data reading. During the write process, the voltage drop introduced by the parasitic resistance may cause insufficient write voltage of the selected cells, resulting in programming errors.…”

Section: Challenges In Device and System Levelsmentioning

confidence: 99%

Essential Characteristics of Memristors for Neuromorphic Computing

Chen

Song

Wang

et al. 2022

Adv Elect Materials

View full text Add to dashboard Cite

increasingly problematic. Unlike the Von Neumann computing platform, the human brain relies on neurons and synapses for storage and computation, which do not have clear boundaries between them. Therefore, nanodevices that mimic synapses, for high-efficiency computing, have been investigated; among these nanodevices, memristors have attracted most attention because of their low power consumption, high integration density, and the ability to simulate synaptic plasticity, which meet the standards of neuromorphic computing. [4] The first report on the resistive switching phenomenon dates back to the 1960s; [5] since early theories were insufficient to explain this phenomenon, research had been done on it. It was not until the memristor was theoretically proposed in 1971, that the mechanisms underpinning the resistive switching became abundant. [6] The first memristor was manufactured by Hewlett-Packard in 2008. [7] Since then, memristors made of diverse materials have been successfully studied, including conductive filament memristors, magnetic tunnel junctions, ferroelectric tunnel junctions, phase-change memristors, and so on (Figure 1). These devices have been used for storage and computing purposes. [8,9] In recent years, there have been many reviews investigating neuromorphic computing from the perspectives of device electrical properties, [9,10] resistive switching materials, [11,12] memristive synapses and neurons, [13] algorithm optimization, [14] and circuit design. [15] Different from the existing literature, we discuss the possibility of achieving brain-like computing from the perspective of memristor technology and review the establishment of spiking neural network neuromorphic computing systems. In this article, we first review the resistive switching mechanisms of different types of memristors and focus on factors, which affect device stability and the corresponding optimization measures that have been applied. Furthermore, we study the stochasticity, power consumption, switching speed, retention, endurance, and other properties of memristors, which are the basis for neuromorphic computing implementations. We then review various memristor-based neural networks and the building of spike neural network neuromorphic computing systems. Finally, we shed light upon the major challenges and offer our perspectives and opinions for memristor-based brainlike computing systems.The memristor is a resistive switch where its resistive state is programable based on the applied voltage or current. Memristive devices are thus capable of storing and computing information simultaneously, breaking the Von Neumann bottleneck. Since the first nanomemristor made by Hewlett-Packard in 2008, advances so far have enabled nanostructured, low-power, high-durability devices that exhibit superior performance over conventional CMOS devices. Herein, the development of memristors based on different physical mechanisms is reviewed. In particular, device stability, integration density, power consumption, switching speed, retention, and e...

show abstract

“…An adaptive programming technique considering parasitic resistance correction was developed, to avoid to use the additional correction circuit or selecting devices. Figure 12(a) show a conventional programming method, where each memristive crossing-point is programmed to corresponding memristance value (Truong 2019). In figure 12(a), for the ith column, the jth input produces the column current, I j,i , represented by the dashed line.…”

Section: Correction Technique Of Parasitic Line Resistance In Memrist...mentioning

confidence: 99%

“…Figure 13 shows the comparison of recognition rate between the conventional programming method and the adaptive programming approach when the parasitic line resistance varies from 0.5 Ω to 3 Ω (Truong 2019).…”

Section: Correction Technique Of Parasitic Line Resistance In Memrist...mentioning

confidence: 99%

Quantization, training, parasitic resistance correction, and programming techniques of memristor-crossbar neural networks for edge intelligence

Nguyen

et al. 2022

Neuromorph. Comput. Eng.

View full text Add to dashboard Cite

In Internet-of-Things (IoT) era, edge intelligence is critical for overcoming the communication and computing energy crisis, which is unavoidable if cloud computing is used exclusively. Memristor crossbars with in-memory computing may be suitable for realizing edge intelligence hardware. They can perform both memory and computing functions, allowing for the development of low-power computing architectures that go beyond the Von Neumann computer. For implementing edge-intelligence hardware with memristor crossbars, in this paper, we review various techniques such as quantization, training, parasitic resistance correction, and low-power crossbar programming, and so on. In particular, memristor crossbars can be considered to realize quantized neural networks with binary and ternary synapses. For preventing memristor defects from degrading edge intelligence performance, chip-in-the-loop training can be useful when training memristor crossbars. Another undesirable effect in memristor crossbars is parasitic resistances such as source, line, and neuron resistance, which worsens as crossbar size increases. Various circuit and software techniques can compensate for parasitic resistances like source, line, and neuron resistance. Finally, we discuss an energy-efficient programming method for updating synaptic weights in memristor crossbars, which is needed for learning the edge devices.

show abstract

A Parasitic Resistance-Adapted Programming Scheme for Memristor Crossbar-Based Neuromorphic Computing Systems

Cited by 5 publications

References 36 publications

Memristors for Neuromorphic Circuits and Artificial Intelligence Applications

Memristors for Neuromorphic Circuits and Artificial Intelligence Applications

Essential Characteristics of Memristors for Neuromorphic Computing

Quantization, training, parasitic resistance correction, and programming techniques of memristor-crossbar neural networks for edge intelligence

Contact Info

Product

Resources

About