Summary

One way to improve the energy efficiency of executing neural networks (NN) is to deploy and execute them on Field Programmable Gate Arrays (FPGA). This allows to execute NNs efficiently "in hardware" while being able to adapt their structure at any time. Due to the limited amount of available resources on embedded FPGAs however NNs have to be optimised so they can be executable on them.

Instead of try to implement generic artificial neurons for a preexisting NN strtucture, we design tailormade neurons, optimised for the strengths and weaknesses of FPGAs. These neurons can then be deployed with very few hardware resources (i.e. by using a limited amount of Lookup Tables or LUTs) and can therefore be realised much more energy efficiently. We showcase the potential for this solution by detecting artifacts in ECG heart data.

Duration

Project start:  01.10.2019

Project end: 31.12.2020

Goal

Using pretrained and preoptimised neural networks we are developing AI solutions that can be executed highly efficiently on FPGAs. We show the potential by detecting artifacts in ECG heart data.

Publications

  • Einhaus, Lukas; Qian, Chao; Ringhofer, Christopher; Schiele, Gregor
    Towards Precomputed 1D-Convolutional Layers for Embedded FPGAs
    In: Machine Learning and Principles and Practice of Knowledge Discovery in Databases: Proceedings, Part I / International Workshops of ECML PKDD 2021, Virtual Event, September 13-17, 2021 / Kamp, Michael; Koprinska, Irena; Bibal, Adrien; Bouadi, Tassadit; Frénay, Benoît; Galárraga, Luis; Oramas, José.; Adilova, Linara; Krishnamurthy, Yamuna; Kang, Bo; Largeron, Christine; Lijffijt, Jefrey; Viard, Tiphaine; Welke, Pascal; Ruocco, Massimiliano; Aune, Erlend; Gallicchio, Claudio; Schiele, Gregor; Pernkopf, Franz; Blott, Michaela; Fröning, Holger; Schindler, Günther; Guidotti, Riccardo; Monreale, Anna; Rinzivillo, Salvatore; Biecek, Przemyslaw; Ntoutsi, Eirini; Pechenizkiy, Mykola; Rosenhahn, Bodo; Buckley, Christopher; Cialfi, Daniela; Lanillos, Pablo; Ramstead, Maxwell; Verbelen, Tim; Ferreira, Pedro M.; Andresini, Giuseppina; Malerba, Donato; Medeiros, Ibéria; Fournier-Viger, Philippe; Nawaz, M. Saqib; Ventura, Sebastian; Sun, Meng; Zhou, Min; Bitetta, Valerio; Bordino, Ilaria; Ferretti, Andrea; Gullo, Francesco; Ponti, Giovanni; Severini, Lorenzo; Ribeiro, Rita; Gama, João; Gavaldà, Ricard; Cooper, Lee; Ghazaleh, Naghmeh; Richiardi, Jonas; Roqueiro, Damian; Saldana Miranda, Diego; Sechidis, Konstantinos; Graça, Guilherme (Hrsg.). Cham: Springer, 2021, S. 327 – 338
    DOI
  • Einhaus, Lukas; Qian, Chao; Ringhofer, Christopher; Schiele, Gregor
    In-Situ Artificial Intelligence for Self- Devices: The Elastic AI Ecosystem (Tutorial)
    In: 2nd IEEE International Conference on Autonomic Computing and Self-Organizing Systems Companion, ACSOS-C 2021: Proceedings / 2nd IEEE International Conference on Autonomic Computing and Self-Organizing Systems Companion, ACSOS-C 2021, Virtual, Washington, 27 September - 1 October 2021. Piscataway: IEEE, 2021, S. 320 – 321
    DOI
  • Burger, Alwyn; Schiele, Gregor; King, David W.
    Developing Action Policies with Q-Learning and Shallow Neural Networks on Reconfigurable Embedded Devices
    In: ACM Transactions on Autonomous and Adaptive Systems (TAAS), Jg. 15, 2021, Nr. 4, 14
    DOI
  • Burger, Alwyn Johannes; King, David W.; Schiele, Gregor
    Reconfigurable Embedded Devices Using Reinforcement Learning to Develop Action-Policies
    In: 2020 IEEE 1st International Conference on Autonomic Computing and Self-Organizing Systems: Proceedings / IEEE 1st International Conference on Autonomic Computing and Self-Organizing Systems (ACSOS), 17-21 August 2020, Washington, DC, USA. Piscataway: IEEE, 2020
    DOI
  • Burger, Alwyn Johannes; Qian, Chao; Schiele, Gregor; Helms, Domenik
    An Embedded CNN Implementation for On-Device ECG Analysis
    In: IEEE Annual Conference on Pervasive Computing and Communications Workshops (PerCom) / PerIoT 2020: The Fourtternational Workshop on Mobile and Pervasive Internet of Thingsh In. Piscataway: IEEE, 2020
    DOI
  • Burger, Alwyn Johannes; Urban, Patrick; Boubin, Jayson; Schiele, Gregor
    An architecture for solving the eigenvalue problem on embedded fpgas
    In: Architecture of Computing Systems – ARCS 2020 / 33rd International Conference on Architecture of Computing Systems; Aachen, Germany; May 25–28, 2020 / Brinkmann, André; Karl, Wolfgang; Lankes, Stefan; Tomforde, Sven; Pionteck, Thilo; Trinitis, Carsten (Hrsg.). Cham: Springer, 2020, S. 32 – 43
    DOI
  • Cichiwskyj, Christopher; Qian, Chao; Schiele, Gregor
    Time to Learn: Temporal Accelerators as an Embedded Deep Neural Network Platform
    In: IoT Streams for Data-Driven Predictive Maintenance and IoT, Edge, and Mobile for Embedded Machine Learning: Second International Workshop, IoT Streams 2020, and First International Workshop, ITEM 2020, Co-located with ECML/PKDD 2020; Ghent, Belgium, September 14-18, 2020; Revised Selected Papers / 2nd International Workshop on IoT Streams for Data-Driven Predictive Maintenance; IoT Streams 2020; September 14-18, 2020, Ghent, Belgium / Gama, João; Pashami, Sepideh; Bifet, Albert; Sayed-Mouchaweh, Moamar; Fröning, Holger; Pernkopf, Franz; Schiele, Gregor; Blott, Michaela (Hrsg.). Cham: Springer, 2020, S. 256 – 267
    DOI