Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 52 of the 60 references it cites were found in a registry or an index, and 8 are not checked yet.

Automating Analogue AI Chip Design with Genetic Search

Open access
Krestinskaya, Olga, Salama, Khaled N., James, Alex P. · 2020 · Advanced Intelligent Systems

Abstract

Optimization of analogue neural circuit designs is one of the most challenging, complicated, time‐consuming, and expensive tasks. Design automation of analogue neuromemristive chips is made difficult by the need to design chips at low cost, ease of scaling, high‐energy efficiency, and small on‐chip area. The rapid progress in edge AI computing applications generates high demand for developing smart sensors. The integration of high‐density analogue computing AI chips as coprocessing units to sensors is gaining popularity. This article proposes a hardware–software codesign framework to speed up and automate the design of analogue neuromemristive chips. This work uses genetic algorithms with objective functions that take into account hardware nonidealities such as limited precision of devices, the device‐to‐device variability, and device failures. The optimized neural architectures and hyperparameters successfully map with the library of relevant neuromemristive analogue hardware blocks. The results demonstrate the advantage of proposed automation to speed up the analogue circuit design of large‐scale neuromemristive networks and reduce overall design costs for AI chips.

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.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1002/aisy.202000075

Cites

52 of 60 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Memristor nonideality affecting the performance of memristive neural networks: a) resistance variation, b) device aging (when several resistive states are removed), and c) endurance and device failure. d) Design automation process.
Figure 1 Memristor nonideality affecting the performance of memristive neural networks: a) resistance variation, b) device aging (when several resistive states are removed), and c) endurance and device failure. d) Design automation process.CC-BY · From the PDF
Figure 2 Comparison of average generation accuracy for ideal neural network and memristive crossbar-based neural network with nonidealities for a) MNIST database and b) Fashion-MNIST database. c) Comparison of average generation accuracy for ideal neural network and memristive crossbar-based neural network with nonidealities for face recognition with AR database, where convolutional layer is fixed and dense layer configuration is selected using genetic search.
Figure 2 Comparison of average generation accuracy for ideal neural network and memristive crossbar-based neural network with nonidealities for a) MNIST database and b) Fashion-MNIST database. c) Comparison of average generation accuracy for ideal neural network and memristive crossbar-based neural network with nonidealities for face recognition with AR database, where convolutional layer is fixed and dense layer configuration is selected using genetic search.CC-BY · From the PDF
Figure 3 a,b) Average and maximum accuracy in a generation and selected network configurations for crossbar-based networks with limited number of stable resistive states L ¼ 2. c,d) Average and maximum accuracy in a generation and selected network configurations for crossbar-based networks with Gaussian memristance variation σ ¼ 25. e,f ) Average and maximum accuracy in a generation and selected network configurations for crossbar-based networks with F ¼ 5% possibility of memristor failure.
Figure 3 a,b) Average and maximum accuracy in a generation and selected network configurations for crossbar-based networks with limited number of stable resistive states L ¼ 2. c,d) Average and maximum accuracy in a generation and selected network configurations for crossbar-based networks with Gaussian memristance variation σ ¼ 25. e,f ) Average and maximum accuracy in a generation and selected network configurations for crossbar-based networks with F ¼ 5% possibility of memristor failure.CC-BY · From the PDF
Image 4
CC-BY · From the PDF
Figure 5 a) Selection of CNN architecture for object recognition for CIFAR-10 dataset with nine hyperparameters. b) Selection of CNN architecture considering the training/optimization algorithm as a network hyperparameter. c) Example of genetic search application for memristive LSTM architecture design.
Figure 5 a) Selection of CNN architecture for object recognition for CIFAR-10 dataset with nine hyperparameters. b) Selection of CNN architecture considering the training/optimization algorithm as a network hyperparameter. c) Example of genetic search application for memristive LSTM architecture design.CC-BY · From the PDF
Figure 6 Hardware implementation of the selected networks from Table 2. After the genetic search, the optimum and most efficient hardware implementation can be selected based on the designer ' s preferences.
Figure 6 Hardware implementation of the selected networks from Table 2. After the genetic search, the optimum and most efficient hardware implementation can be selected based on the designer ' s preferences.CC-BY · From the PDF

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.

A few free checks a day · papers already verified open instantly · results are public

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?

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

Show all 48 references

Research works printed but not linked: 21

  1. Bergstra J., 2012, J. Mach. Learn. Res., 13, 281
  2. J.Snoek O.Rippel K.Swersky R.Kiros N.Satish N.Sundaram M.Patwary M.Prabhat R.Adams inInt. Conf. Machine Learning PMLR Lille France2015 pp.2171–2180.
  3. D. B.Fogel inBCEC World Scientific Press Sweden1997 pp.1–11.
  4. H.Xiao K.Rasul R.Vollgraf arXiv preprint arXiv:1708.077472017.
  5. Vourkas I., 2016, IJUC, 12, 265
  6. Ielmini D., 2019, Nanotechnology, 9, 092001
  7. B.Yan M.Liu Y.Chen K.Chakrabarty H.Li inIEEE Int. Electron Devices Meeting (IEDM) IEEE San Francisco CA2019 14–5. · doi:10.1109/iedm19573.2019.8993562
  8. C.Liu M.Hu J. P.Strachan H.Li in54th ACM/EDAC/IEEE Design Automation Conf. (DAC) IEEE Austin TX2017 pp.1–6.
  9. N.Amos C.Stewart P.Bhat C.Cretsinger E.Won W.Dharmaratna H.Prosper inComputing in High Energy Physics’ 95: CHEP’95 World Scientific Singapore1996 pp.215–219. · doi:10.1142/9789814447188_0038
  10. R.Bardenet M.Brendel B.Kégl M.Sebag inInt. Conf. Machine Learning JMLR Atlanta GA2013 pp.199–207.
  11. M.Pelikan D. E.Goldberg E.Cantú-Paz inProc. of the Genetic and Evolutionary Computation Conf. GECCO-99 Vol.1 Morgan Kaufmann Publishers Orlando FL1999 pp.525–532.
  12. Li L., 2017, J. Mach. Learn. Res., 18, 6765
  13. M.Jaderberg V.Dalibard S.Osindero W. M.Czarnecki J.Donahue A.Razavi O.Vinyals T.Green I.Dunning K.Simonyan C.Fernando arXiv preprint arXiv:1711.098462017.
  14. A.Klein S.Falkner S.Bartels P.Hennig F.Hutter inInt. Conf. on Artificial Intelligence and Statistics (AISTATS) JMLR Fort Lauderdale FL2017 pp.528–536.
  15. S.Falkner A.Klein F.Hutter inInt. Conf. on Machine Learning PMLR Stockholm Sweden2018 pp.1437–1446.
  16. X.Yao inProc. of IEEE Int. Symp. on Parallel Algorithms Architecture Synthesis IEEE Aizu-Wakamatsu1997 pp.282–291.
  17. D. P.Kingma J.Ba inInt. Conf. on Learning Representations (ICLR) San Diego CA2015.
  18. Duchi J., 2011, J. Mac. Learn. Res., 12, 2121
  19. I.Sutskever J.Martens G.Dahl G.Hinton inInt. Conf. on Machine Learning JMLR Atlanta GA2013 pp.1139–1147.
  20. M. D.Zeiler arXiv preprint arXiv:1212.57012012.
  21. T.Dozat inInt. Conf. on Learning Representations (ICLR) Caribe Hilton San Juan Puerto Rico2016.

Cited by

See all 14

Metadata

Authors
Krestinskaya, O., Salama, K. N., & James, A. P.
Affiliations
  • Nazarbayev University (KZ)
  • King Abdullah University of Science and Technology (SA)
  • Indian Institute of Information Technology and Management, Kerala (IN)
Corresponding author
James, Alex P.
Venue
Advanced Intelligent Systems
Publisher
Wiley
Published
2020-06-25
Volume
2
Issue
8
ISSN
2640-4567 (ISSN-L)
OpenAlex
W3037092152
Cited by
14
References
48
Citations by year
2026: 1 · 2025: 1 · 2024: 4 · 2023: 2 · 2022: 4 · 2021: 1 · 2020: 1
FWCI
0.69
Type
journal article
Language
en
Open access
Locations
  • Advanced Intelligent Systems (publishedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Nazarbayev University Repository (Nazarbayev University) (submittedVersion, cc-by-nc-sa)
  • King Abdullah University of Science and Technology Repository (King Abdullah University of Science and Technology) (submittedVersion, other-oa)
Funders
King Abdullah University of Science and Technology
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 › Computer Science › Artificial Intelligence › Evolutionary Algorithms and Applications
Keywords
genetic algorithm, design automation, hardware-software co-design, edge AI, smart sensors, device-to-device variation, device failure, neural architecture search

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: Automating Analogue AI Chip Design with Genetic Search. https://app.cochat.ai/papers/10.1002/aisy.202000075
The paper
Krestinskaya, O., Salama, K. N., & James, A. P. (2020). Automating Analogue AI Chip Design with Genetic Search. Advanced Intelligent Systems. https://doi.org/10.1002/aisy.202000075

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.