- author = "Michal Pinos and Vojtech Mrazek and Lukas Sekanina",
- title = "Evolutionary Neural Architecture Search Supporting Approximate Multipliers",
- booktitle = "EuroGP 2021: Proceedings of the 24th European Conference on Genetic Programming",
- year = "2021",
- editor = "Ting Hu and Nuno Lourenco and Eric Medvet",
- series = "LNCS",
- volume = "12691",
- publisher = "Springer Verlag",
- address = "Virtual Event",
- pages = "82--97",
- month = "7-9 " # apr,
- organisation = "EvoStar, Species",
- note = "Best paper nomination",
- keywords = "genetic algorithms, genetic programming, Cartesian genetic programming, CGP, ANN, approximate computing, convolutional neural network, Neuroevolution, Energy Efficiency, IOT, TensorFlow, NSGA-II",
- isbn13 = "978-3-030-72811-3",
-
URL = "
https://arxiv.org/abs/2101.11883",
-
DOI = "
doi:10.1007/978-3-030-72812-0_6",
- size = "16 pages",
- abstract = "There is a growing interest in automated neural architecture search (NAS) methods. They are employed to routinely deliver high-quality neural network architectures for various challenging data sets and reduce the designer's effort. The NAS methods using multiobjective evolutionary algorithms are especially useful when the objective is not only to minimize the network error but also to minimize the number of parameters (weights) or power consumption of the inference phase. We propose a multi-objective NAS method based on Cartesian genetic programming for evolving convolutional neural networks (CNN). The method allows approximate operations to be used in CNNs to reduce the power consumption of a target hardware implementation. During the NAS process, a suitable CNN architecture is evolved together with approximate multipliers to deliver the best trade-offs between the accuracy, network size, and power consumption. The most suitable approximate multipliers are automatically selected from a library of approximate multipliers. Evolved CNNs are compared with common human-created CNNs of a similar complexity on the CIFAR-10 benchmark problem.",
-
notes = "FIT, Brno University of Technology.
Deep ANN image processing energy expensive on internet of things micro device. Evolutionary NAS. CGP NSGA2 Tailor CNN for hardware. cf \cite{Suganuma:2017:GECCO} Count multiply and accumulate (MAC) operations in CNN convolution layers. Use 8bit operations instead of 32bit floating. EvoApproxLib \cite{7926993} TFAapprox Neuroevolution. Pop=8, 10+1 gens. ResNet, ALWANN.
http://www.evostar.org/2021/eurogp/ Part of \cite{Hu:2021:GP} EuroGP'2021 held in conjunction with EvoCOP2021, EvoMusArt2021 and EvoApplications2021",
