Position innerhalb des Seitenbaumes

Start
Institut
Team
Neugebauer, Florian

Florian Neugebauer

Herr Dr. rer. nat.

Wissenschaftlicher Mitarbeiter
Institut für Technische Informatik
Hardwareorientierte Informatik

Kontakt

+49 711 685 88389
+49 711 685 88288
E-Mail

Pfaffenwaldring 47
D-70569 Stuttgart
Deutschland
Raum: 3.172

Publikationen

2022
1. On the Impact of Hardware Timing Errors on Stochastic Computing based Neural Networks. Florian Neugebauer; Stefan Holst and Ilia Polian. In 2022 IEEE European Test Symposium (ETS), 2022, pp. 1–6. DOI: https://doi.org/10.1109/ETS54262.2022.9810429
  - BibTeX
  - DOI
  BibTeX
  @inproceedings{9810429, abteilung = {hocos}, author = {Neugebauer, Florian and Holst, Stefan and Polian, Ilia}, booktitle = {2022 IEEE European Test Symposium (ETS)}, doi = {10.1109/ETS54262.2022.9810429}, pages = {1-6}, title = {On the Impact of Hardware Timing Errors on Stochastic Computing based Neural Networks}, year = 2022 }
  DOI
  10.1109/ETS54262.2022.9810429
2020
1. Retraining and Regularization to Optimize Neural Networks for Stochastic Computing. Junseok Oh; Florian Neugebauer; Ilia Polian and John Hayes. In 2020 IEEE Computer Society Annual Symposium on VLSI (ISVLSI), 2020, pp. 246–251. DOI: https://doi.org/10.1109/ISVLSI49217.2020.00052
  Zusammenfassung
  Stochastic computing (SC) is a promising computation technique for applications with huge numbers of individually simple operations. One of the main application targets for SC is the design of convolutional neural networks (CNNs) due to their heavy reliance on multiply-accumulate operations that have compact and power-efficient SC realizations. We present a training optimization method that can improve the accuracy of SC-based CNNs significantly. Using regularization techniques in combination with a newly developed retraining algorithm, we enforce sparsity, reduce the number of required MUX-based additions, and thus minimize undesired downscaling. Our results show that the proposed regularized training procedure (RTP) can reduce the computation time of SC-based NNs by a factor of four without sacrificing classification accuracy while also reducing area in hardware implementations. These findings suggest that SC can be successfully applied to more powerful CNNs than previously thought possible, thus extending the range of potential uses of SC.
  BibTeX
  @inproceedings{9154942, abstract = {Stochastic computing (SC) is a promising computation technique for applications with huge numbers of individually simple operations. One of the main application targets for SC is the design of convolutional neural networks (CNNs) due to their heavy reliance on multiply-accumulate operations that have compact and power-efficient SC realizations. We present a training optimization method that can improve the accuracy of SC-based CNNs significantly. Using regularization techniques in combination with a newly developed retraining algorithm, we enforce sparsity, reduce the number of required MUX-based additions, and thus minimize undesired downscaling. Our results show that the proposed regularized training procedure (RTP) can reduce the computation time of SC-based NNs by a factor of four without sacrificing classification accuracy while also reducing area in hardware implementations. These findings suggest that SC can be successfully applied to more powerful CNNs than previously thought possible, thus extending the range of potential uses of SC.}, abteilung = {hocos}, author = {{Oh}, Junseok and {Neugebauer}, Florian and {Polian}, Ilia and {Hayes}, John}, booktitle = {2020 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)}, doi = {10.1109/ISVLSI49217.2020.00052}, isbn = {978-1-7281-5775-7}, issn = {2159-3477}, month = {07}, pages = {246-251}, title = {Retraining and Regularization to Optimize Neural Networks for Stochastic Computing}, year = 2020 }
  DOI
  10.1109/ISVLSI49217.2020.00052
2. Hardware-based fast real-time image classification with stochastic computing. Ponana Kelettira Muthappa; Florian Neugebauer; Ilia Polian and John P. Hayes. In Proceedings of the 38th IEEE International Conference on Computer Design (ICCD’20), 2020.
  - BibTeX
  BibTeX
  @inproceedings{kelettiramuthappa2020hardwarebased, abteilung = {hocos}, author = {Kelettira Muthappa, Ponana and Neugebauer, Florian and Polian, Ilia and Hayes, John P.}, booktitle = {Proceedings of the 38th IEEE International Conference on Computer Design (ICCD'20)}, isbn = {978-1-7281-9710-4}, month = {10}, title = {Hardware-based fast real-time image classification with stochastic computing}, year = 2020 }
2019
1. On the maximum function in stochastic computing. Florian Neugebauer; Ilia Polian and John P. Hayes. In Proceedings of the 16th ACM International Conference on Computing Frontiers, CF 2019, Alghero, Italy, April 30 - May 2, 2019., 2019, pp. 59--66. DOI: https://doi.org/10.1145/3310273.3323050
  BibTeX
  @inproceedings{DBLP:conf/cf/NeugebauerPH19, abteilung = {hocos}, author = {Neugebauer, Florian and Polian, Ilia and Hayes, John P.}, bibsource = {dblp computer science bibliography, https://dblp.org}, booktitle = {Proceedings of the 16th {ACM} International Conference on Computing Frontiers, {CF} 2019, Alghero, Italy, April 30 - May 2, 2019.}, crossref = {DBLP:conf/cf/2019}, doi = {10.1145/3310273.3323050}, pages = {59--66}, title = {On the maximum function in stochastic computing}, url = {https://doi.org/10.1145/3310273.3323050}, year = 2019 }
  Link
  https://doi.org/10.1145/3310273.3323050
  DOI
  10.1145/3310273.3323050
2. On the Limits of Stochastic Computing. Florian Neugebauer; Ilia Polian and John P. Hayes. In 2019 IEEE International Conference on Rebooting Computing, ICRC, 2019, pp. 98--105. DOI: https://doi.org/10.1109/ICRC.2019.8914706
  BibTeX
  @inproceedings{DBLP:conf/icrc/NeugebauerPH19, abteilung = {hocos}, author = {Neugebauer, Florian and Polian, Ilia and Hayes, John P.}, bibsource = {dblp computer science bibliography, https://dblp.org}, booktitle = {2019 {IEEE} International Conference on Rebooting Computing, {ICRC}}, day = {6--8}, doi = {10.1109/ICRC.2019.8914706}, month = {11}, pages = {98--105}, publisher = {{IEEE}}, title = {On the Limits of Stochastic Computing}, url = {https://doi.org/10.1109/ICRC.2019.8914706}, year = 2019 }
  Link
  https://doi.org/10.1109/ICRC.2019.8914706
  DOI
  10.1109/ICRC.2019.8914706
2018
1. S-box-based random number generation for stochastic computing. Florian Neugebauer; Ilia Polian and John P. Hayes. Microprocessors and Microsystems - Embedded Hardware Design 61, (2018), pp. 316--326. DOI: https://doi.org/10.1016/j.micpro.2018.06.009
  BibTeX
  @article{DBLP:journals/mam/NeugebauerPH18, abteilung = {hocos}, author = {Neugebauer, Florian and Polian, Ilia and Hayes, John P.}, bibsource = {dblp computer science bibliography, https://dblp.org}, doi = {10.1016/j.micpro.2018.06.009}, journal = {Microprocessors and Microsystems - Embedded Hardware Design}, pages = {316--326}, title = {S-box-based random number generation for stochastic computing}, url = {https://doi.org/10.1016/j.micpro.2018.06.009}, volume = 61, year = 2018 }
  Link
  https://doi.org/10.1016/j.micpro.2018.06.009
  DOI
  10.1016/j.micpro.2018.06.009
2. Framework for Quantifying and Managing Accuracy in Stochastic Circuit Design. Florian Neugebauer; Ilia Polian and John P. Hayes. JETC 14, 2 (2018), pp. 31:1--31:21. DOI: https://doi.org/10.1145/3183345
  BibTeX
  @article{DBLP:journals/jetc/NeugebauerPH18, abteilung = {hocos}, author = {Neugebauer, Florian and Polian, Ilia and Hayes, John P.}, bibsource = {dblp computer science bibliography, https://dblp.org}, doi = {10.1145/3183345}, journal = {{JETC}}, number = 2, pages = {31:1--31:21}, title = {Framework for Quantifying and Managing Accuracy in Stochastic Circuit Design}, url = {https://doi.org/10.1145/3183345}, volume = 14, year = 2018 }
  Link
  https://doi.org/10.1145/3183345
  DOI
  10.1145/3183345
2017
1. Framework for quantifying and managing accuracy in stochastic circuit design. Florian Neugebauer; Ilia Polian and John P. Hayes. In Design, Automation & Test in Europe Conference & Exhibition, DATE 2017, Lausanne, Switzerland, March 27-31, 2017, 2017, pp. 1--6. DOI: https://doi.org/10.23919/DATE.2017.7926949
  BibTeX
  @inproceedings{DBLP:conf/date/NeugebauerPH17, abteilung = {hocos}, author = {Neugebauer, Florian and Polian, Ilia and Hayes, John P.}, bibsource = {dblp computer science bibliography, https://dblp.org}, booktitle = {Design, Automation {\&} Test in Europe Conference {\&} Exhibition, {DATE} 2017, Lausanne, Switzerland, March 27-31, 2017}, crossref = {DBLP:conf/date/2017}, doi = {10.23919/DATE.2017.7926949}, pages = {1--6}, title = {Framework for quantifying and managing accuracy in stochastic circuit design}, url = {https://doi.org/10.23919/DATE.2017.7926949}, year = 2017 }
  Link
  https://doi.org/10.23919/DATE.2017.7926949
  DOI
  10.23919/DATE.2017.7926949
2. Building a Better Random Number Generator for Stochastic Computing. Florian Neugebauer; Ilia Polian and John P. Hayes. In Euromicro Conference on Digital System Design, DSD 2017, Vienna, Austria, August 30 - Sept. 1, 2017, 2017, pp. 1--8. DOI: https://doi.org/10.1109/DSD.2017.29
  BibTeX
  @inproceedings{DBLP:conf/dsd/NeugebauerPH17, abteilung = {hocos}, author = {Neugebauer, Florian and Polian, Ilia and Hayes, John P.}, bibsource = {dblp computer science bibliography, https://dblp.org}, booktitle = {Euromicro Conference on Digital System Design, {DSD} 2017, Vienna, Austria, August 30 - Sept. 1, 2017}, crossref = {DBLP:conf/dsd/2017}, doi = {10.1109/DSD.2017.29}, pages = {1--8}, title = {Building a Better Random Number Generator for Stochastic Computing}, url = {https://doi.org/10.1109/DSD.2017.29}, year = 2017 }
  Link
  https://doi.org/10.1109/DSD.2017.29
  DOI
  10.1109/DSD.2017.29

Florian Neugebauer

Kontakt

Publikationen

2022

BibTeX

DOI

2020

Zusammenfassung

BibTeX

DOI

BibTeX

2019

BibTeX

Link

DOI

BibTeX

Link

DOI

2018

BibTeX

Link

DOI

BibTeX

Link

DOI

2017

BibTeX

Link

DOI

BibTeX

Link

DOI

Zielgruppe

Formalia

Services

Organisation

Florian Neugebauer

Kontakt

Publikationen

2022

BibTeX

DOI

2020

Zusammenfassung

BibTeX

DOI

BibTeX

2019

BibTeX

Link

DOI

BibTeX

Link

DOI

2018

BibTeX

Link

DOI

BibTeX

Link

DOI

2017

BibTeX

Link

DOI

BibTeX

Link

DOI

So erreichen Sie uns

Zielgruppe

Formalia

Services

Organisation