The DOI matches its Crossref record; 67 of the 72 references it cites were found in a registry or an index.
Mixed-Signal Computing for Deep Neural Network Inference
Abstract
Modern deep neural networks (DNNs) require billions of multiply-accumulate operations per inference. Given that these computations demand relatively low precision, it is feasible to consider analog computing, which can be more efficient than digital in the low-SNR regime. This overview article investigates the potential of mixed analog/digital computing approaches in the context of modern DNN processor architectures, which are typically limited by memory access. We discuss how memory-like and in-memory compute fabrics may help alleviate this bottleneck and derive asymptotic efficiency limits at the processing array level. It is shown that single-digit fJ/op energy efficiencies are feasible for 4-bit mixed-signal arithmetic. In this analysis, special consideration is given to the SNR and amortization requirements of the analog-digital interfaces. In addition, we consider the pros and cons for a variety of implementation styles and highlight the challenge of retaining high compute efficiency for a complete DNN accelerator design.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Interestingly, the VMM operation —often understood as multiply and accumulate (MAC) routine—can be implemented using a crossbar array of memory elements. Those memory devices could be either charge-based memories as well as resistance-based memories25,74.”
From Hardware implementation of memristor-based artificial neural networks (2024) · cites this paper as [74]
Citations
Cites
Very Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148
Gradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805
MobileNetV2: Inverted Residuals and Linear Bottlenecks
Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., & Chen, L.-C. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00474 · Cited by 28,755
Efficient Processing of Deep Neural Networks: A Tutorial and Survey
Sze, V., Chen, Y.-H., Yang, T.-J., & Emer, J. S. (2017) · Proceedings of the IEEE · doi:10.1109/jproc.2017.2761740 · Cited by 4,215 · cited 3×
“In particular, we focus on convolution layers, which tend to dominate the size and workload of modern DNNs with millions of parameters [6].”
BackgroundCourbariaux, M., Hubara, I., Soudry, D., El-Yaniv, R., & Bengio, Y. (2016) · arXiv · doi:10.48550/arxiv.1602.02830 · Cited by 2,106
ISAAC: a convolutional neural network accelerator with in-situ analog arithmetic in crossbars
Shafiee, A., Nag, A., Muralimanohar, N., Balasubramonian, R., Strachan, J. P., Hu, M., Williams, R. S., & Srikumar, V. (2016) · ACM SIGARCH Computer Architecture News · doi:10.1145/3007787.3001139 · Cited by 1,536
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 · cited 2×
“Progress in this field heavily relies on hardware-algorithm co-design [5] and has led to a variety of insightful design principles for custom digital DNN chips [6]–[9].”
BackgroundShiDianNao: shifting vision processing closer to the sensor
Du, Z., Fasthuber, R., Chen, T., Ienne, P., Li, L., Luo, T., Feng, X., Chen, Y., & Temam, O. (2015) · Proceedings of the 42nd Annual International Symposium on Computer Architecture · doi:10.1145/2749469.2750389 · Cited by 933
A fully integrated reprogrammable memristor–CMOS system for efficient multiply–accumulate operations
Cai, F., Correll, J. M., Lee, S. H., Lim, Y., Bothra, V., Zhang, Z., Flynn, M. P., & Lu, W. D. (2019) · Nature Electronics · doi:10.1038/s41928-019-0270-x · Cited by 728
A new golden age for computer architecture
Hennessy, J. L., & Patterson, D. A. (2019) · Communications of the ACM · doi:10.1145/3282307 · Cited by 690
Research works printed but not linked: 3
- guo, 2020, et al Neural Network Accelerator Comparison
- murmann, 2020, ADC Performance Survey 1997–2020
- goodfellow, 2016, Deep Learning
Cited by
Memristive technologies for data storage, computation, encryption, and radio-frequency communication
Lanza, M., Sebastian, A., Lu, W. D., Le Gallo, M., Chang, M.-F., Akinwande, D., Puglisi, F. M., Alshareef, H. N., Liu, M., & Roldan, J. B. (2022) · Science · doi:10.1126/science.abj9979 · Cited by 708
“However, with the potential for substantially smaller areal footprint and even 3D integration, memristive IMC is expected to have a substantial density advantage even at very advanced CMOS nodes (87).”
BackgroundHardware 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
“Thosememory devicescould beeither charge-based memories as well as resistance-based memories 25,74 .”
Experimentally realized in situ backpropagation for deep learning in photonic neural networks
Pai, S., Sun, Z., Hughes, T. W., Park, T., Bartlett, B., Williamson, I. A. D., Minkov, M., Milanizadeh, M., Abebe, N., Morichetti, F., et al. (2023) · Science · doi:10.1126/science.ade8450 · Cited by 331
Recent Advances and Future Prospects for Memristive Materials, Devices, and Systems
Song, M.-K., Kang, J.-H., Zhang, X., Ji, W., Ascoli, A., Messaris, I., Demirkol, A. S., Dong, B., Aggarwal, S., Wan, W., et al. (2023) · ACS Nano · doi:10.1021/acsnano.3c03505 · Cited by 300
Le Gallo, M., Khaddam-Aljameh, R., Stanisavljevic, M., Vasilopoulos, A., Kersting, B., Dazzi, M., Karunaratne, G., Brändli, M., Singh, A., Müller, S. M., et al. (2023) · Nature Electronics · doi:10.1038/s41928-023-01010-1 · Cited by 310
“Finally, the weight values of each submatrix are converted into conductance values using equation (1), with Gmax being determined on a per-core basis.”
Method
Metadata
- Authors
- Murmann, B.
- Affiliations
- Stanford University (US)
- Corresponding author
- Murmann, Boris
- ORCID
- Murmann, Boris
- Venue
- IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Published
- 2021-01
- Volume
- 29
- Issue
- 1
- Pages
- 3-13
- ISSN
- 1063-8210 (ISSN-L)
- 1557-9999
- OpenAlex
- W3120016950
- Cited by
- 160
- References
- 72
- Citations by year
- 2026: 24 · 2025: 42 · 2024: 37 · 2023: 25 · 2022: 21 · 2021: 10
- FWCI
- 4.90
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Transactions on Very Large Scale Integration (VLSI) Systems (publishedVersion)
- Funders
- Stanford’s SystemX Alliance
- National Science Foundation (1937294)
- Topics
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Memory and Neural Computing
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Analog and Mixed-Signal Circuit Design
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Low-power high-performance VLSI design
- Keywords
- analog computing, deep neural network inference, DNN accelerators, in-memory computing, memory access bottleneck, energy efficiency, analog-digital interface, multiply-accumulate operations
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: Mixed-Signal Computing for Deep Neural Network Inference. https://app.cochat.ai/papers/10.1109/tvlsi.2020.3020286
- The paper
- Murmann, B. (2021). Mixed-Signal Computing for Deep Neural Network Inference. IEEE Transactions on Very Large Scale Integration (VLSI) Systems. https://doi.org/10.1109/tvlsi.2020.3020286
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.