On the Accuracy of Analog Neural Network Inference Accelerators
Open accessThe DOI matches its Crossref record; 68 of the 71 references it cites were found in a registry or an index.
Abstract
Specialized accelerators have recently garnered attention as a method to reduce the power consumption of neural network inference. A promising category of accelerators utilizes nonvolatile memory arrays to both store weights and perform in situ analog computation inside the array. While prior work has explored the design space of analog accelerators to optimize performance and energy efficiency, there is seldom a rigorous evaluation of the accuracy of these accelerators. This work shows how architectural design decisions, particularly in mapping neural network parameters to analog memory cells, influence inference accuracy. When evaluated using ResNet50 on ImageNet, the resilience of the system to analog non-idealities—cell programming errors, analog-to-digital converter resolution, and array parasitic resistances—all improve when analog quantities in the hardware are made proportional to the numerical values in the network. Moreover, contrary to the assumptions of prior work, nearly equivalent resilience to cell imprecision can be achieved by fully storing weights as analog quantities, rather than spreading weight bits across multiple devices, often referred to as bit slicing. By exploiting proportionality, analog system designers have the freedom to match the precision of the hardware to the needs of the algorithm, rather than attempting to guarantee the same level of precision in the intermediate results as an equivalent digital accelerator. This ultimately results in an analog accelerator that is more accurate, more robust to analog errors, and more energy-efficient.
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.
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?
The cited source doesn’t support the claim as written.
In the citing paper
“ADCs with at least 8-bit resolution are necessary to achieve high (>90%) classification accuracy in a ResNET50-1.5 ANN used to classify the ImageNET157 database”
From Hardware implementation of memristor-based artificial neural networks (2024) · cites this paper as [157]
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measured
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
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
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
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 · cited 1×
“The sensitivity to cell errors (described more fully in Section 5) differs dramatically for networks trained on three datasets—ImageNet [15], CIFAR10 [36], and MNIST [37]—with ImageNet being by far the most sensitive.”
BackgroundRethinking the Inception Architecture for Computer Vision
Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., & Wojna, Z. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.308 · Cited by 32,138
Learning Multiple Layers of Features from Tiny Images
Krizhevsky, A. (2024) · doi:10.57702/zp44cu3g · Cited by 23,395
MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., & Adam, H. (2017) · arXiv · doi:10.48550/arxiv.1704.04861 · Cited by 19,015
Han, S., Mao, H., & Dally, W. J. (2015) · arXiv · doi:10.48550/arxiv.1510.00149 · Cited by 4,662
In-Datacenter Performance Analysis of a Tensor Processing Unit
Jouppi, N. P., Young, C., Patil, N., Patterson, D., Agrawal, G., Bajwa, R., Bates, S., Bhatia, S., Boden, N., Borchers, A., et al. (2017) · Proceedings of the 44th Annual International Symposium on Computer Architecture · doi:10.1145/3079856.3080246 · Cited by 4,537
Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference
Jacob, B., Kligys, S., Chen, B., Zhu, M., Tang, M., Howard, A., Adam, H., & Kalenichenko, D. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00286 · Cited by 4,114
End-to-end arguments in system design
Saltzer, J. H., Reed, D. P., & Clark, D. D. (1984) · ACM Transactions on Computer Systems · doi:10.1145/357401.357402 · Cited by 2,538 · cited 1×
“[53], this robustness allows hardware requirements (such as the array size and the analog-to-digital converter resolution) to be relaxed while still ensuring high end-to-end accuracy of an application neural network.”
Background
Research works printed but not linked: 8
- chollet, 2015, Keras
- murmann, 2020, ADC Performance Survey 1997–2020
- agarwal, 2017, CrossSim
- 2019, MLPerf Inference V0 5 NVIDIA-Optimized Implementations for Open Division
- 2019, MLPerf Inference ImageNet Calibration Set
- 2019, MLPerf Inference Benchmark Suite
- lecun, 2021, The MNIST Database of Handwritten Digits
- hayakawa, 2015, Highly reliable TaOx ReRAM with centralized filament for 28-nm embedded application, Proc Symp VLSI Technol, 14t
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
Andrulis, T., Emer, J., & Sze, V. (2023) · Proceedings - International Symposium on Computer Architecture/Proceedings · doi:10.1145/3579371.3589062 · Cited by 70
“Offset encoding provides noise resistance [73], and Center+Offset increases bit sparsity and decreases noise.”
BackgroundMethodInfluentialXiao, T. P., Feinberg, B., Bennett, C. H., Agrawal, V. S., Saxena, P., Prabhakar, V., Ramkumar, K., Medu, H., Raghavan, V., Chettuvetty, R., et al. (2022) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2021.3134313 · Cited by 46
Self-citationCiMLoop: A Flexible, Accurate, and Fast Compute-In-Memory Modeling Tool
Andrulis, T., Emer, J., & Sze, V. (2024) · IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS) · doi:10.1109/ispass61541.2024.00012 · Cited by 60
“2) Representation: How data values appear is determined by how the hardware represents operands [47].”
BackgroundAnalog Multiply-Accumulate Cell With Multi-Bit Resolution for All-Analog AI Inference Accelerators
Nägele, R., Finkbeiner, J., Stadtlander, V., Grözing, M., & Berroth, M. (2023) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2023.3268728 · Cited by 27
“property of neural networks supports this statement [9].”
BackgroundInfluential
Metadata
- Authors
- Xiao, T. P., Feinberg, B., Bennett, C. H., Prabhakar, V., Saxena, P., Agrawal, V., Agarwal, S., & Marinella, M. J.
- Affiliations
- Sandia National Laboratories California (US)
- Infineon Technologies (United States) (US)
- Arizona State University (US)
- ORCID
- Venue
- IEEE Circuits and Systems Magazine
- Published
- 2022
- Volume
- 22
- Issue
- 4
- Pages
- 26-48
- ISSN
- 0163-6812 (ISSN-L)
- 1531-636X
- 1558-0830
- OpenAlex
- W3198899759
- Cited by
- 76
- References
- 71
- Citations by year
- 2026: 18 · 2025: 30 · 2024: 18 · 2023: 8 · 2022: 2
- FWCI
- 3.97
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Circuits and Systems Magazine (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University) (cc-by)
- Funders
- U.S. Department of Energy (DENA0003525, NA0003525)
- National Nuclear Security Administration (NA0003525, DENA-0003525)
- Sandia National Laboratories (NA0003525, DE‐NA,0003525)
- Topics
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Memory and Neural Computing
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Low-power high-performance VLSI design
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Analog and Mixed-Signal Circuit Design
- Keywords
- in-sensor computing, inference accuracy, bit slicing, analog-to-digital converter resolution, parasitic resistance, coding errors, ResNet50, energy efficiency
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: On the Accuracy of Analog Neural Network Inference Accelerators. https://app.cochat.ai/papers/10.1109/mcas.2022.3214409
- The paper
- Xiao, T. P., Feinberg, B., Bennett, C. H., Prabhakar, V., Saxena, P., Agrawal, V., Agarwal, S., & Marinella, M. J. (2022). On the Accuracy of Analog Neural Network Inference Accelerators. IEEE Circuits and Systems Magazine. https://doi.org/10.1109/mcas.2022.3214409
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.