RxNN: A Framework for Evaluating Deep Neural Networks on Resistive Crossbars
Open accessThe DOI matches its Crossref record; 9 of the 9 references it cites were found in a registry or an index.
Abstract
Resistive crossbars designed with nonvolatile memory devices have emerged as promising building blocks for deep neural network (DNN) hardware, due to their ability to compactly and efficiently realize vector-matrix multiplication (VMM), the dominant computational kernel in DNNs. However, a key challenge with resistive crossbars is that they suffer from a range of device and circuit level nonidealities, such as driver resistance, sensing resistance, sneak paths, interconnect parasitics, nonlinearities in the peripheral circuits, stochastic write operations, and process variations. These nonidealities can lead to errors in VMMs, eventually degrading the DNN's accuracy. It is therefore critical to study the impact of crossbar nonidealities on the accuracy of large-scale DNNs (with millions of neurons and billions of synaptic connections). However, this is challenging because the existing device and circuit models are too slow to use in application-level evaluations. We present RxNN, a fast and accurate simulation framework to evaluate large-scale DNNs on resistive crossbar systems. RxNN splits and maps the computations involved in each DNN layer into crossbar operations, and evaluates them using a fast crossbar model (FCM) that accurately captures the errors arising due to crossbar nonidealities while being four-to-five orders of magnitude faster than circuit simulation. FCM models a crossbar-based VMM operation using three stages-nonlinear models for the input and output peripheral circuits (digital-to-analog and analog-to-digital converters), and an equivalent nonideal conductance matrix for the core crossbar array. We implement RxNN by extending the Caffe machine learning framework and use it to evaluate a suite of six large-scale DNNs developed for the ImageNet Challenge (ILSVRC). Our experiments reveal that resistive crossbar nonidealities can lead to significant accuracy degradations (9.6%-32%) for these large-scale DNNs. To the best of our knowledge, this article is the first quantitative evaluation of the accuracy of large-scale DNNs on resistive crossbar-based hardware. We also demonstrate that RxNN enables fast model-in-the-loop retraining of DNNs to partially mitigate the accuracy degradation.
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
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1512.03385 · Cited by 4,196
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
OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks
Sermanet, P., Eigen, D., Zhang, X., Mathieu, M., Fergus, R., & LeCun, Y. (2013) · arXiv · doi:10.48550/arxiv.1312.6229 · Cited by 2,680
Deep Learning with Limited Numerical Precision
Gupta, S., Agrawal, A., Gopalakrishnan, K., & Narayanan, P. (2015) · arXiv · doi:10.48550/arxiv.1502.02551 · Cited by 1,364
Gokmen, T., & Vlasov, Y. (2016) · Frontiers in Neuroscience · doi:10.3389/fnins.2016.00333 · Cited by 491
Spin-Transfer Torque Magnetic Memory as a Stochastic Memristive Synapse for Neuromorphic Systems
Vincent, A. F., Larroque, J., Locatelli, N., Ben Romdhane, N., Bichler, O., Gamrat, C., Zhao, W. S., Klein, J.-O., Galdin-Retailleau, S., & Querlioz, D. (2015) · IEEE Transactions on Biomedical Circuits and Systems · doi:10.1109/tbcas.2015.2414423 · Cited by 360
Accelerator-friendly neural-network training: Learning variations and defects in RRAM crossbar
Chen, L., Li, J., Chen, Y., Deng, Q., Shen, J., Liang, X., & Jiang, L. (2017) · Design, Automation & Test in Europe Conference & Exhibition (DATE), 2017 · doi:10.23919/date.2017.7926952 · Cited by 245 · cited 1×
“These include: 1) (re-)training [16]–[21]; 2) optimized weight to conductance conversion [14]; 3) rank clipping to reduce the effects of nonidealities by lowering cross-bar dimensions [25]; 4) schemes to alleviate the effect of hard failures [26]; and 5) hardware solutions to address low-voltage…”
BackgroundTraining Deep Convolutional Neural Networks with Resistive Cross-Point Devices
Gokmen, T., Onen, M., & Haensch, W. (2017) · Frontiers in Neuroscience · doi:10.3389/fnins.2017.00538 · Cited by 162
Cited by
Model Compression and Hardware Acceleration for Neural Networks: A Comprehensive Survey
Deng, B. L., Li, G., Han, S., Shi, L., & Xie, Y. (2020) · Proceedings of the IEEE · doi:10.1109/jproc.2020.2976475 · Cited by 996
“Considering the crossbar mapping and the low-bit limitation during the training of models can Adapted from [296]. help recover the application accuracy [290], [291].”
BackgroundEmbedded Devices for Neuromorphic Time-Series Assessment
Mazumder, A. N., Hosseini, M., & Mohsenin, T. (2022) · Research Explorer (The University of Manchester) · doi:10.13016/m2ksy0-wxr3 · Cited by 652
Compute-in-Memory Chips for Deep Learning: Recent Trends and Prospects
Yu, S., Jiang, H., Huang, S., Peng, X., & Lu, A. (2021) · IEEE Circuits and Systems Magazine · doi:10.1109/mcas.2021.3092533 · Cited by 391
Rasch, M. J., Mackin, C., Le Gallo, M., Chen, A., Fasoli, A., Odermatt, F., Li, N., Nandakumar, S. R., Narayanan, P., Tsai, H., et al. (2023) · Nature Communications · doi:10.1038/s41467-023-40770-4 · Cited by 213
“Some aspectsofour AIMCcrossbar model have been investigated individually in earlier studies, such as the effect of ADC/DAC quantization, IR-drop, and general read noise 55 – 57 , as well as data-dependent long-term noise 38 .”
BackgroundMethodA Flexible and Fast PyTorch Toolkit for Simulating Training and Inference on Analog Crossbar Arrays
Rasch, M. J., Moreda, D., Gokmen, T., Le Gallo, M., Carta, F., Goldberg, C., El Maghraoui, K., Sebastian, A., & Narayanan, V. (2021) · 2021 IEEE 3rd International Conference on Artificial Intelligence Circuits and Systems (AICAS) · doi:10.1109/aicas51828.2021.9458494 · Cited by 160
“While there exists a number of current software packages that are dedicated to simulating AI workloads on non-volatile memory elements, notably MLP-NEUROSIM [6] and RXNN [7], they are usually not tied to modern AI frameworks, such as PYTORCH [8].”
Metadata
- Authors
- Jain, S., Sengupta, A., Roy, K., & Raghunathan, A.
- Venue
- IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Published
- 2021-02
- Volume
- 40
- Issue
- 2
- Pages
- 326-338
- ISSN
- 0278-0070 (ISSN-L)
- 1937-4151
- OpenAlex
- W2909715355
- Cited by
- 137
- References
- 9
- FWCI
- 6.58
- Type
- journal article
- Language
- en
- Open access
- green
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: RxNN: A Framework for Evaluating Deep Neural Networks on Resistive Crossbars. https://app.cochat.ai/papers/10.1109/tcad.2020.3000185
- The paper
- Jain, S., Sengupta, A., Roy, K., & Raghunathan, A. (2021). RxNN: A Framework for Evaluating Deep Neural Networks on Resistive Crossbars. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems. https://doi.org/10.1109/tcad.2020.3000185
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.