Simulation of Inference Accuracy Using Realistic RRAM Devices
Open accessThe DOI matches its Crossref record; 12 of the 12 references it cites were found in a registry or an index.
Abstract
Resistive Random Access Memory (RRAM) is a promising technology for power efficient hardware in applications of artificial intelligence (AI) and machine learning (ML) implemented in non-von Neumann architectures. However, there is an unanswered question if the device non-idealities preclude the use of RRAM devices in this potentially disruptive technology. Here we investigate the question for the case of inference. Using experimental results from silicon oxide (SiOx) RRAM devices, that we use as proxies for physical weights, we demonstrate that acceptable accuracies in classification of handwritten digits (MNIST data set) can be achieved using non-ideal devices. We find that, for this test, the ratio of the high- and low-resistance device states is a crucial determinant of classification accuracy, with ~96.8% accuracy achievable for ratios >3, compared to ~97.3% accuracy achieved with ideal weights. Further, we investigate the effects of a finite number of discrete resistance states, sub-100% device yield, devices stuck at one of the resistance states, current/voltage non-linearities, programming non-linearities and device-to-device variability. Detailed analysis of the effects of the non-idealities will better inform the need for the optimization of particular device properties.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Nanoscale Memristor Device as Synapse in Neuromorphic Systems
Jo, S. H., Chang, T., Ebong, I., Bhadviya, B. B., Mazumder, P., & Lu, W. (2010) · Nano Letters · doi:10.1021/nl904092h · Cited by 4,174 · cited 1×
“As a result, different learning laws have been realized using these devices, e.g., spike-timing-dependent plasticity for spiking neural networks (Jo et al., 2010; Serrano-Gotarredona et al., 2013; Yu et al., 2013a; Chang et al., 2016).”
BackgroundTraining and operation of an integrated neuromorphic network based on metal-oxide memristors
Prezioso, M., Merrikh-Bayat, F., Hoskins, B. D., Adam, G. C., Likharev, K. K., & Strukov, D. B. (2015) · Nature · doi:10.1038/nature14441 · Cited by 3,001 · cited 1×
“Some impressive proof of concept work has been conducted, achieving supervised (Prezioso et al., 2015) or unsupervised learning (Serb et al., 2016) using RRAM-based crossbars as PNNs (Ielmini, 2018).”
BackgroundNeuromorphic electronic systems
Mead, C. (1990) · Proceedings of the IEEE · doi:10.1109/5.58356 · Cited by 2,368 · cited 1×
“Closely related is a neuromorphic approach (Mead, 1989, 1990; Poon and Zhou, 2011), that draws inspiration from the human brain, which is a remarkable power-efficient system—many orders of magnitude better than the current most power-efficient CMOS systems.”
BackgroundNanoelectronic programmable synapses based on phase change materials for brain-inspired computing
Kuzum, D., Jeyasingh, R. G. D., Lee, B., & Wong, H.-S. P. (2012) · Nano Letters · doi:10.1021/nl201040y · Cited by 1,267 · cited 4×
“This method was used to achieve state-of-the-art accuracy by augmenting PCM devices with CMOS circuitry to deal with non-idealities (Ambrogio et al., 2018).”
BackgroundInfluentialNeuro-Inspired Computing With Emerging Nonvolatile Memorys
Yu, S. (2018) · Proceedings of the IEEE · doi:10.1109/jproc.2018.2790840 · Cited by 1,170 · cited 1×
“This is the simplest approach being considered by others (Yu, 2018) because it minimizes the complexity of required circuitry for mapping.”
BackgroundEquivalent-accuracy accelerated neural-network training using analogue memory
Ambrogio, S., Narayanan, P., Tsai, H., Shelby, R. M., Boybat, I., di Nolfo, C., Sidler, S., Giordano, M., Bodini, M., Farinha, N. C. P., et al. (2018) · Nature · doi:10.1038/s41586-018-0180-5 · Cited by 1,151
Short-Term Memory to Long-Term Memory Transition in a Nanoscale Memristor
Chang, T., Jo, S.-H., & Lu, W. (2011) · ACS Nano · doi:10.1021/nn202983n · Cited by 1,058 · cited 1×
“…phase-change memories (PCM) (Kuzum et al., 2011; Nandakumar et al., 2018; Sebastian et al., 2018) and resistive random-access memories (RRAM) (Chang et al., 2011; Yu et al., 2013b; Serb et al., 2016; Stathopoulos et al., 2017; Pi et al., 2019), both of which may be considered to be…”
BackgroundBurr, G. W., Shelby, R. M., Sidler, S., di Nolfo, C., Jang, J., Boybat, I., Shenoy, R. S., Narayanan, P., Virwani, K., Giacometti, E. U., et al. (2015) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2015.2439635 · Cited by 936 · cited 1×
“…voltage-current characteristics, non-linearity of resistance modulation with voltage pulses) and system levels (sneak currents and high resistances of interconnections in crossbar arrays) cannot be ignored and can lead to reduction in computational accuracy (Burr et al., 2015; Chai et al., 2018).”
BackgroundSub-nanosecond switching of a tantalum oxide memristor
Torrezan, A. C., Strachan, J. P., Medeiros-Ribeiro, G., & Williams, R. S. (2011) · Nanotechnology · doi:10.1088/0957-4484/22/48/485203 · Cited by 701 · cited 1×
“Resistive RAM technology is based on simple two terminal (metal-oxide-metal) nanodevices whose resistance can be repeatedly varied, with low operational energy and very high levels of integration (Torrezan et al., 2011; Mehonic et al., 2012).”
BackgroundGokmen, T., & Vlasov, Y. (2016) · Frontiers in Neuroscience · doi:10.3389/fnins.2016.00333 · Cited by 491
Cited by
Hardware implementation of memristor-based artificial neural networks
Aguirre, F., Sebastian, A., Le Gallo, M., Song, W., Wang, T., Yang, J. J., Lu, W., Chang, M.-F., Ielmini, D., Yang, Y., et al. (2024) · Nature Communications · doi:10.1038/s41467-024-45670-9 · Cited by 564
Self-citationMehonic, A., Sebastian, A., Rajendran, B., Simeone, O., Vasilaki, E., & Kenyon, A. J. (2020) · Advanced Intelligent Systems · doi:10.1002/aisy.202000085 · Cited by 331
Self-citationStandards for the Characterization of Endurance in Resistive Switching Devices
Lanza, M., Waser, R., Ielmini, D., Yang, J. J., Goux, L., Suñe, J., Kenyon, A. J., Mehonic, A., Spiga, S., Rana, V., et al. (2021) · ACS Nano · doi:10.1021/acsnano.1c06980 · Cited by 262
Self-citationNonideality‐Aware Training for Accurate and Robust Low‐Power Memristive Neural Networks
Joksas, D., Wang, E., Barmpatsalos, N., Ng, W. H., Kenyon, A. J., Constantinides, G. A., & Mehonic, A. (2022) · Advanced Science · doi:10.1002/advs.202105784 · Cited by 39
“Such nonidealities include stuck devices, deviceto-device (D2D) and cycle-to-cycle variability, drift in resistance states, line resistance, I-V nonlinearity, and others [10].”
BackgroundSelf-citationAguirre, F. L., Pazos, S. M., Palumbo, F., Sune, J., & Miranda, E. (2020) · IEEE Access · doi:10.1109/access.2020.3035638 · Cited by 36
“Second, Python-based approaches and similar [21], [22] allow incorporating realistic RRAM models, but normally ignore the parasitic CPA effects or do not account for the peripheral circuitry.”
Method
Metadata
- Authors
- Mehonic, A., Joksas, D., Ng, W. H., Buckwell, M., & Kenyon, A. J.
- Venue
- Frontiers in Neuroscience
- Publisher
- Frontiers Media SA
- Published
- 2019-06-12
- Volume
- 13
- Pages
- 593-593
- ISSN
- 1662-453X (ISSN-L)
- OpenAlex
- W2952615984
- Cited by
- 64
- References
- 12
- FWCI
- 3.61
- Type
- journal article
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Simulation of Inference Accuracy Using Realistic RRAM Devices. https://app.cochat.ai/papers/10.3389/fnins.2019.00593
- The paper
- Mehonic, A., Joksas, D., Ng, W. H., Buckwell, M., & Kenyon, A. J. (2019). Simulation of Inference Accuracy Using Realistic RRAM Devices. Frontiers in Neuroscience. https://doi.org/10.3389/fnins.2019.00593
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.