{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,5]],"date-time":"2026-02-05T06:38:32Z","timestamp":1770273512619,"version":"3.49.0"},"reference-count":76,"publisher":"Association for Computing Machinery (ACM)","issue":"4","license":[{"start":{"date-parts":[[2020,8,4]],"date-time":"2020-08-04T00:00:00Z","timestamp":1596499200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"name":"Swiss State Secretariat for Education, Research and Innovation","award":["16.0159"],"award-info":[{"award-number":["16.0159"]}]},{"name":"European Union's Horizon 2020 research and innovation programme","award":["732204 (Bonseyes)"],"award-info":[{"award-number":["732204 (Bonseyes)"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Internet Things"],"published-print":{"date-parts":[[2020,11,30]]},"abstract":"<jats:p>\n            Next generation of embedded Information and Communication Technology (ICT) systems are interconnected and collaborative systems able to perform autonomous tasks. The remarkable expansion of the embedded ICT market, together with the rise and breakthroughs of Artificial Intelligence (AI), have put the focus on the\n            <jats:italic>Edge<\/jats:italic>\n            as it stands as one of the keys for the next technological revolution: the seamless integration of AI in our daily life. However, training and deployment of custom AI solutions on embedded devices require a fine-grained integration of data, algorithms, and tools to achieve high accuracy and overcome functional and non-functional requirements. Such integration requires a high level of expertise that becomes a real bottleneck for small and medium enterprises wanting to deploy AI solutions on the\n            <jats:italic>Edge<\/jats:italic>\n            , which, ultimately, slows down the adoption of AI on applications in our daily life.\n          <\/jats:p>\n          <jats:p>\n            In this work, we present a modular AI pipeline as an integrating framework to bring data, algorithms, and deployment tools together. By removing the integration barriers and lowering the required expertise, we can interconnect the different stages of particular tools and provide a modular end-to-end development of AI products for embedded devices. Our AI pipeline consists of four modular main steps:\n            <jats:italic>(i)<\/jats:italic>\n            data ingestion,\n            <jats:italic>(ii)<\/jats:italic>\n            model training,\n            <jats:italic>(iii)<\/jats:italic>\n            deployment optimization, and\n            <jats:italic>(iv)<\/jats:italic>\n            the IoT hub integration. To show the effectiveness of our pipeline, we provide examples of different AI applications during each of the steps. Besides, we integrate our deployment framework, Low-Power Deep Neural Network (LPDNN), into the AI pipeline and present its lightweight architecture and deployment capabilities for embedded devices. Finally, we demonstrate the results of the AI pipeline by showing the deployment of several AI applications such as keyword spotting, image classification, and object detection on a set of well-known embedded platforms, where LPDNN consistently outperforms all other popular deployment frameworks.\n          <\/jats:p>","DOI":"10.1145\/3403572","type":"journal-article","created":{"date-parts":[[2020,8,4]],"date-time":"2020-08-04T16:53:18Z","timestamp":1596559998000},"page":"1-25","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":20,"title":["Bonseyes AI Pipeline\u2014Bringing AI to You"],"prefix":"10.1145","volume":"1","author":[{"given":"Miguel De","family":"Prado","sequence":"first","affiliation":[{"name":"Haute Ecole Arc Ingenierie; HES-SO\/Integrated Systems Lab, ETH Zurich, Switzerland"}]},{"given":"Jing","family":"Su","sequence":"additional","affiliation":[{"name":"School of Computer Science and Statistics, Trinity College Dublin, Ireland"}]},{"given":"Rabia","family":"Saeed","sequence":"additional","affiliation":[{"name":"Haute Ecole Arc Ingenierie; HES-SO, Switzerland"}]},{"given":"Lorenzo","family":"Keller","sequence":"additional","affiliation":[{"name":"Nviso, Switzerland"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5092-8275","authenticated-orcid":false,"given":"Noelia","family":"Vallez","sequence":"additional","affiliation":[{"name":"Universidad de Castilla\u2014La Mancha, Spain"}]},{"given":"Andrew","family":"Anderson","sequence":"additional","affiliation":[{"name":"School of Computer Science and Statistics, Trinity College Dublin, Ireland"}]},{"given":"David","family":"Gregg","sequence":"additional","affiliation":[{"name":"School of Computer Science and Statistics, Trinity College Dublin, Ireland"}]},{"given":"Luca","family":"Benini","sequence":"additional","affiliation":[{"name":"Integrated Systems Lab, ETH Zurich, Switzerland"}]},{"given":"Tim","family":"Llewellynn","sequence":"additional","affiliation":[{"name":"Nviso, Switzerland"}]},{"given":"Nabil","family":"Ouerhani","sequence":"additional","affiliation":[{"name":"Haute Ecole Arc Ingenierie; HES-SO, Switzerland"}]},{"given":"Rozenn","family":"Dahyot","sequence":"additional","affiliation":[{"name":"School of Computer Science 8 Statistics, Trinity College Dublin, Ireland"}]},{"given":"Nuria","family":"Pazos","sequence":"additional","affiliation":[{"name":"Haute Ecole Arc Ingenierie; HES-SO, Switzerland"}]}],"member":"320","published-online":{"date-parts":[[2020,8,4]]},"reference":[{"key":"e_1_2_1_1_1","unstructured":"2017. Google Speech Commands Dataset. Retrieved from https:\/\/ai.googleblog.com\/2017\/08\/launching-speech-commands-dataset.html.  2017. Google Speech Commands Dataset. Retrieved from https:\/\/ai.googleblog.com\/2017\/08\/launching-speech-commands-dataset.html."},{"key":"e_1_2_1_2_1","unstructured":"2017. The ONNX Project. Retrieved from https:\/\/github.com\/onnx\/onnx.  2017. The ONNX Project. Retrieved from https:\/\/github.com\/onnx\/onnx."},{"key":"e_1_2_1_3_1","unstructured":"2017. Open Neural Network Exchange (ONNX). Retrieved from https:\/\/onnx.ai\/.  2017. Open Neural Network Exchange (ONNX). Retrieved from https:\/\/onnx.ai\/."},{"key":"e_1_2_1_4_1","unstructured":"2018. Arm Compute Library. Retrieved from https:\/\/developer.arm.com\/ip-products\/processors\/machine-learning\/compute-library.  2018. Arm Compute Library. Retrieved from https:\/\/developer.arm.com\/ip-products\/processors\/machine-learning\/compute-library."},{"key":"e_1_2_1_5_1","unstructured":"2018. Bonseyes Official Caffe 1.0 Version. Retrieved from https:\/\/github.com\/bonseyes\/caffe-jacinto.  2018. Bonseyes Official Caffe 1.0 Version. Retrieved from https:\/\/github.com\/bonseyes\/caffe-jacinto."},{"key":"e_1_2_1_6_1","unstructured":"2018. Caffe. Retrieved from http:\/\/caffe.berkeleyvision.org\/.  2018. Caffe. Retrieved from http:\/\/caffe.berkeleyvision.org\/."},{"key":"e_1_2_1_7_1","unstructured":"2018. Caffe-SSD. Retrieved from https:\/\/github.com\/weiliu89\/caffe.  2018. Caffe-SSD. Retrieved from https:\/\/github.com\/weiliu89\/caffe."},{"key":"e_1_2_1_8_1","unstructured":"2018. Discover the Power of Artificial Intelligence to Drive ICT Innovation. Retrieved from https:\/\/news.itu.int\/discover-the-power-of-artificial-intelligence-to-drive-ict-innovation-in-the-first-issue-of-the-itu-journal\/.  2018. Discover the Power of Artificial Intelligence to Drive ICT Innovation. Retrieved from https:\/\/news.itu.int\/discover-the-power-of-artificial-intelligence-to-drive-ict-innovation-in-the-first-issue-of-the-itu-journal\/."},{"key":"e_1_2_1_9_1","unstructured":"2018. Docker. Retrieved from http:\/\/www.docker.com.  2018. Docker. Retrieved from http:\/\/www.docker.com."},{"key":"e_1_2_1_10_1","unstructured":"2018. FI-ware Project. Retrieved from https:\/\/www.fiware.org\/.  2018. FI-ware Project. Retrieved from https:\/\/www.fiware.org\/."},{"key":"e_1_2_1_11_1","unstructured":"2018. Machine Learning Fragmentation Is Slowing Us Down: There Is a Solution. Retrieved from https:\/\/www.cmswire.com\/digital-experience\/machine-learning-fragmentation-is-slowing-us-down-there-is-a-solution\/.  2018. Machine Learning Fragmentation Is Slowing Us Down: There Is a Solution. Retrieved from https:\/\/www.cmswire.com\/digital-experience\/machine-learning-fragmentation-is-slowing-us-down-there-is-a-solution\/."},{"key":"e_1_2_1_12_1","unstructured":"2018. Scientists Can\u2019t Replicate AI Studies. That\u2019s Bad News. Retrieved from https:\/\/futurism.com\/scientists-cant-replicate-ai-studies.  2018. Scientists Can\u2019t Replicate AI Studies. That\u2019s Bad News. Retrieved from https:\/\/futurism.com\/scientists-cant-replicate-ai-studies."},{"key":"e_1_2_1_13_1","unstructured":"2019. Amazon Machine Learning on AWS. Retrieved from https:\/\/aws.amazon.com\/machine-learning.  2019. Amazon Machine Learning on AWS. Retrieved from https:\/\/aws.amazon.com\/machine-learning."},{"key":"e_1_2_1_14_1","unstructured":"2019. Amazon SageMaker. Retrieved from https:\/\/aws.amazon.com\/sagemaker\/.  2019. Amazon SageMaker. Retrieved from https:\/\/aws.amazon.com\/sagemaker\/."},{"key":"e_1_2_1_15_1","unstructured":"2019. Android Neural Networks API (NNAPI). Retrieved from https:\/\/developer.android.com\/ndk\/guides\/neuralnetworks.  2019. Android Neural Networks API (NNAPI). Retrieved from https:\/\/developer.android.com\/ndk\/guides\/neuralnetworks."},{"key":"e_1_2_1_16_1","unstructured":"2019. Apple AI. Retrieved from https:\/\/www.zdnet.com\/article\/apple-says-artificial-intelligence-and-machine-learning-critical-area-as-it-promotes-ai-chief\/.  2019. Apple AI. Retrieved from https:\/\/www.zdnet.com\/article\/apple-says-artificial-intelligence-and-machine-learning-critical-area-as-it-promotes-ai-chief\/."},{"key":"e_1_2_1_17_1","unstructured":"2019. CAFFE2. Retrieved from https:\/\/caffe2.ai\/.  2019. CAFFE2. Retrieved from https:\/\/caffe2.ai\/."},{"key":"e_1_2_1_18_1","unstructured":"2019. CAFFE2FB. Retrieved from https:\/\/research.fb.com\/downloads\/caffe2\/.  2019. CAFFE2FB. Retrieved from https:\/\/research.fb.com\/downloads\/caffe2\/."},{"key":"e_1_2_1_19_1","unstructured":"2019. CNTK: The Microsoft Cognitive Toolkit. Retrieved from https:\/\/docs.microsoft.com\/en-us\/cognitive-toolkit.  2019. CNTK: The Microsoft Cognitive Toolkit. Retrieved from https:\/\/docs.microsoft.com\/en-us\/cognitive-toolkit."},{"key":"e_1_2_1_20_1","unstructured":"2019. Converting Trained Models to Core ML. Retrieved from https:\/\/developer.apple.com\/documentation\/coreml\/converting_trained_models_to_core_ml  2019. Converting Trained Models to Core ML. Retrieved from https:\/\/developer.apple.com\/documentation\/coreml\/converting_trained_models_to_core_ml"},{"key":"e_1_2_1_21_1","unstructured":"2019. Core ML: Integrate Machine Learning Models Into Your App. Retrieved from https:\/\/developer.apple.com\/documentation\/coreml.  2019. Core ML: Integrate Machine Learning Models Into Your App. Retrieved from https:\/\/developer.apple.com\/documentation\/coreml."},{"key":"e_1_2_1_22_1","unstructured":"2019. Folding of Bnorm Into Convolution. Retrieved from https:\/\/tehnokv.com\/posts\/fusing-batchnorm-and-conv\/.  2019. Folding of Bnorm Into Convolution. Retrieved from https:\/\/tehnokv.com\/posts\/fusing-batchnorm-and-conv\/."},{"key":"e_1_2_1_23_1","unstructured":"2019. Google AI Platform. Retrieved from https:\/\/cloud.google.com\/ai-platform\/.  2019. Google AI Platform. Retrieved from https:\/\/cloud.google.com\/ai-platform\/."},{"key":"e_1_2_1_24_1","unstructured":"2019. Google AI Tools. Retrieved from https:\/\/ai.google\/tools\/.  2019. Google AI Tools. Retrieved from https:\/\/ai.google\/tools\/."},{"key":"e_1_2_1_25_1","unstructured":"2019. Google ML. Retrieved from https:\/\/cloud.google.com\/products\/machine-learning\/.  2019. Google ML. Retrieved from https:\/\/cloud.google.com\/products\/machine-learning\/."},{"key":"e_1_2_1_26_1","unstructured":"2019. Greengrass. Retrieved from https:\/\/aws.amazon.com\/greengrass\/.  2019. Greengrass. Retrieved from https:\/\/aws.amazon.com\/greengrass\/."},{"key":"e_1_2_1_27_1","unstructured":"2019. Greengrass Region. Retrieved from https:\/\/aws.amazon.com\/about-aws\/global-infrastructure\/regional-product-services\/.  2019. Greengrass Region. Retrieved from https:\/\/aws.amazon.com\/about-aws\/global-infrastructure\/regional-product-services\/."},{"key":"e_1_2_1_28_1","unstructured":"2019. ImageNet. Retrieved from http:\/\/www.image-net.org.  2019. ImageNet. Retrieved from http:\/\/www.image-net.org."},{"key":"e_1_2_1_29_1","unstructured":"2019. Intel OpenVINO Toolkit. Retrieved from https:\/\/docs.openvinotoolkit.org.  2019. Intel OpenVINO Toolkit. Retrieved from https:\/\/docs.openvinotoolkit.org."},{"key":"e_1_2_1_30_1","unstructured":"2019. Keras: The Python Deep Learning Library. Retrieved from https:\/\/keras.io.  2019. Keras: The Python Deep Learning Library. Retrieved from https:\/\/keras.io."},{"key":"e_1_2_1_31_1","unstructured":"2019. Lead in AI. Retrieved from https:\/\/www.forbes.com\/sites\/danielaraya\/2019\/01\/01\/who-will-lead-in-the-age-of-artificial-intelligence\/#4f3a15aa6f95.  2019. Lead in AI. Retrieved from https:\/\/www.forbes.com\/sites\/danielaraya\/2019\/01\/01\/who-will-lead-in-the-age-of-artificial-intelligence\/#4f3a15aa6f95."},{"key":"e_1_2_1_32_1","unstructured":"2019. Microsoft Azure. Retrieved from https:\/\/azure.microsoft.com\/en-us\/services\/machine-learning\/.  2019. Microsoft Azure. Retrieved from https:\/\/azure.microsoft.com\/en-us\/services\/machine-learning\/."},{"key":"e_1_2_1_33_1","unstructured":"2019. Microsoft Azure IoT Edge. Retrieved from https:\/\/azure.microsoft.com\/en-in\/services\/iot-edge\/.  2019. Microsoft Azure IoT Edge. Retrieved from https:\/\/azure.microsoft.com\/en-in\/services\/iot-edge\/."},{"key":"e_1_2_1_34_1","unstructured":"2019. Mobile Neural Network (MNN): A Lightweight Deep Neural Network Inference Engine. Retrieved from https:\/\/github.com\/alibaba\/MNN.  2019. Mobile Neural Network (MNN): A Lightweight Deep Neural Network Inference Engine. Retrieved from https:\/\/github.com\/alibaba\/MNN."},{"key":"e_1_2_1_35_1","unstructured":"2019. Model Interpretability in Azure Machine Learning Service. Retrieved from https:\/\/docs.microsoft.com\/en-us\/azure\/machine-learning\/service\/how-to-machine-learning-interpretability.  2019. Model Interpretability in Azure Machine Learning Service. Retrieved from https:\/\/docs.microsoft.com\/en-us\/azure\/machine-learning\/service\/how-to-machine-learning-interpretability."},{"key":"e_1_2_1_36_1","unstructured":"2019. MXNet: A Flexible and Efficient Library for Deep Learning. Retrieved from https:\/\/mxnet.apache.org.  2019. MXNet: A Flexible and Efficient Library for Deep Learning. Retrieved from https:\/\/mxnet.apache.org."},{"key":"e_1_2_1_37_1","unstructured":"2019. NCNN: A High-Performance Neural Network Inference Framework Optimized for the Mobile Platform.Retrieved from https:\/\/github.com\/Tencent\/ncnn.  2019. NCNN: A High-Performance Neural Network Inference Framework Optimized for the Mobile Platform.Retrieved from https:\/\/github.com\/Tencent\/ncnn."},{"key":"e_1_2_1_38_1","unstructured":"2019. NNPACK. Retrieved from https:\/\/github.com\/Maratyszcza\/NNPACK.  2019. NNPACK. Retrieved from https:\/\/github.com\/Maratyszcza\/NNPACK."},{"key":"e_1_2_1_39_1","unstructured":"2019. Nvidia Jetson Nano. Retrieved from https:\/\/www.nvidia.com\/en-us\/autonomous-machines\/embedded-systems\/jetson-nano\/.  2019. Nvidia Jetson Nano. Retrieved from https:\/\/www.nvidia.com\/en-us\/autonomous-machines\/embedded-systems\/jetson-nano\/."},{"key":"e_1_2_1_40_1","unstructured":"2019. Nvidia Jetson Xavier. Retrieved from https:\/\/developer.nvidia.com\/embedded\/jetson-agx-xavier-developer-kit.  2019. Nvidia Jetson Xavier. Retrieved from https:\/\/developer.nvidia.com\/embedded\/jetson-agx-xavier-developer-kit."},{"key":"e_1_2_1_41_1","unstructured":"2019. Nvidia Realeased of PyTorch. Retrieved from https:\/\/docs.nvidia.com\/deeplearning\/frameworks\/pytorch-release-notes\/overview.html#overview.  2019. Nvidia Realeased of PyTorch. Retrieved from https:\/\/docs.nvidia.com\/deeplearning\/frameworks\/pytorch-release-notes\/overview.html#overview."},{"key":"e_1_2_1_42_1","unstructured":"2019. PyTorch: From Research to Production. Retrieved from https:\/\/pytorch.org.  2019. PyTorch: From Research to Production. Retrieved from https:\/\/pytorch.org."},{"key":"e_1_2_1_43_1","unstructured":"2019. Quantization Analysis Tool. Retrieved from https:\/\/github.com\/BUG1989\/caffe-int8-convert-tools.  2019. Quantization Analysis Tool. Retrieved from https:\/\/github.com\/BUG1989\/caffe-int8-convert-tools."},{"key":"e_1_2_1_44_1","unstructured":"2019. Raspberry Pi3. Retrieved from https:\/\/www.raspberrypi.org\/products\/raspberry-pi-3-model-b-plus\/.  2019. Raspberry Pi3. Retrieved from https:\/\/www.raspberrypi.org\/products\/raspberry-pi-3-model-b-plus\/."},{"key":"e_1_2_1_45_1","unstructured":"2019. Raspberry Pi4. Retrieved from https:\/\/www.raspberrypi.org\/products\/raspberry-pi-4-model-b\/.  2019. Raspberry Pi4. Retrieved from https:\/\/www.raspberrypi.org\/products\/raspberry-pi-4-model-b\/."},{"key":"e_1_2_1_46_1","unstructured":"2019. Renesas e-AI. Retrieved from https:\/\/www.renesas.com\/eu\/en\/solutions\/key-technology\/e-ai.html.  2019. Renesas e-AI. Retrieved from https:\/\/www.renesas.com\/eu\/en\/solutions\/key-technology\/e-ai.html."},{"key":"e_1_2_1_47_1","unstructured":"2019. Tengine: A Lite High-performance and Modular Inference Engine for Embedded Device. Retrieved from https:\/\/github.com\/OAID\/Tengine.  2019. Tengine: A Lite High-performance and Modular Inference Engine for Embedded Device. Retrieved from https:\/\/github.com\/OAID\/Tengine."},{"key":"e_1_2_1_48_1","unstructured":"2019. TensorFlow: An End-to-end Open Source Machine Learning Platform. Retrieved from https:\/\/www.tensorflow.org.  2019. TensorFlow: An End-to-end Open Source Machine Learning Platform. Retrieved from https:\/\/www.tensorflow.org."},{"key":"e_1_2_1_49_1","unstructured":"2019. TensorFlow Lite: Deploy Machine Learning Models on Mobile and IoT Devices. Retrieved from https:\/\/www.tensorflow.org\/lite.  2019. TensorFlow Lite: Deploy Machine Learning Models on Mobile and IoT Devices. Retrieved from https:\/\/www.tensorflow.org\/lite."},{"key":"e_1_2_1_50_1","unstructured":"2019. TensorRT. Retrieved from https:\/\/developer.nvidia.com\/tensorrt.  2019. TensorRT. Retrieved from https:\/\/developer.nvidia.com\/tensorrt."},{"key":"e_1_2_1_51_1","unstructured":"2019. TI-DL. Retrieved from https:\/\/training.ti.com\/texas-instruments-deep-learning-tidl-overview.  2019. TI-DL. Retrieved from https:\/\/training.ti.com\/texas-instruments-deep-learning-tidl-overview."},{"key":"e_1_2_1_52_1","unstructured":"2019. Torch: A Scientific Computing Framework for LuaJIT. Retrieved from http:\/\/torch.ch.  2019. Torch: A Scientific Computing Framework for LuaJIT. Retrieved from http:\/\/torch.ch."},{"key":"e_1_2_1_53_1","volume-title":"Proceedings of the 2019 International Conference on High Performance Computing and Simulation. IEEE.","author":"Anderson Andrew","year":"2019","unstructured":"Andrew Anderson , Jing Su , Rozenn Dahyot , and David Gregg . 2019 . Performance-oriented neural architecture search . In Proceedings of the 2019 International Conference on High Performance Computing and Simulation. IEEE. Andrew Anderson, Jing Su, Rozenn Dahyot, and David Gregg. 2019. Performance-oriented neural architecture search. In Proceedings of the 2019 International Conference on High Performance Computing and Simulation. IEEE."},{"key":"e_1_2_1_54_1","unstructured":"James S. Bergstra R\u00e9mi Bardenet Yoshua Bengio and Bal\u00e1zs K\u00e9gl. 2011. Algorithms for hyper-parameter optimization. In Advances in Neural Information Processing Systems. 2546--2554.  James S. Bergstra R\u00e9mi Bardenet Yoshua Bengio and Bal\u00e1zs K\u00e9gl. 2011. Algorithms for hyper-parameter optimization. In Advances in Neural Information Processing Systems. 2546--2554."},{"key":"e_1_2_1_55_1","volume-title":"Retrieved","author":"Microsoft NNI","year":"2019","unstructured":"Microsoft NNI contributors. 2019 . An Open Source AutoML Toolkit for Neural Architecture Search and Hyper-parameter Tuning . Retrieved May 27, 2019 from https:\/\/github.com\/Microsoft\/nni. Microsoft NNI contributors. 2019. An Open Source AutoML Toolkit for Neural Architecture Search and Hyper-parameter Tuning. Retrieved May 27, 2019 from https:\/\/github.com\/Microsoft\/nni."},{"key":"e_1_2_1_56_1","volume-title":"Proceedings of the 15th ACM International Conference on Computing Frontiers. ACM, 36--44","author":"de Prado Miguel","year":"2018","unstructured":"Miguel de Prado , Maurizio Denna , Luca Benini , and Nuria Pazos . 2018 . QUENN: QUantization engine for low-power neural networks . In Proceedings of the 15th ACM International Conference on Computing Frontiers. ACM, 36--44 . Miguel de Prado, Maurizio Denna, Luca Benini, and Nuria Pazos. 2018. QUENN: QUantization engine for low-power neural networks. In Proceedings of the 15th ACM International Conference on Computing Frontiers. ACM, 36--44."},{"key":"e_1_2_1_57_1","volume-title":"Learning to infer: RL-based search for DNN primitive selection on heterogeneous embedded systems. arXiv preprint arXiv:1811.07315","author":"de Prado Miguel","year":"2018","unstructured":"Miguel de Prado , Nuria Pazos , and Luca Benini . 2018. Learning to infer: RL-based search for DNN primitive selection on heterogeneous embedded systems. arXiv preprint arXiv:1811.07315 ( 2018 ). Miguel de Prado, Nuria Pazos, and Luca Benini. 2018. Learning to infer: RL-based search for DNN primitive selection on heterogeneous embedded systems. arXiv preprint arXiv:1811.07315 (2018)."},{"key":"e_1_2_1_58_1","volume-title":"Jan Hendrik Metzen, and Frank Hutter","author":"Elsken Thomas","year":"2018","unstructured":"Thomas Elsken , Jan Hendrik Metzen, and Frank Hutter . 2018 . Neural architecture search: A survey. arXiv e-prints, Article arXiv:1808.05377 (Aug 2018), arXiv:1808.05377 pages. arxiv:stat.ML\/1808.05377 Thomas Elsken, Jan Hendrik Metzen, and Frank Hutter. 2018. Neural architecture search: A survey. arXiv e-prints, Article arXiv:1808.05377 (Aug 2018), arXiv:1808.05377 pages. arxiv:stat.ML\/1808.05377"},{"key":"e_1_2_1_59_1","volume-title":"Proceedings of the Computing Frontiers Conference. ACM, 299--304","author":"T. Llewellynn","year":"2017","unstructured":"T. Llewellynn et al. 2017 . BONSEYES: Platform for open development of systems of artificial intelligence . In Proceedings of the Computing Frontiers Conference. ACM, 299--304 . T. Llewellynn et al. 2017. BONSEYES: Platform for open development of systems of artificial intelligence. In Proceedings of the Computing Frontiers Conference. ACM, 299--304."},{"key":"e_1_2_1_60_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2013.6638947"},{"key":"e_1_2_1_61_1","volume-title":"Proceedings of the 2017 IEEE High Performance Extreme Computing Conference (HPEC). IEEE, 1--7.","author":"Ho Nhut-Minh","year":"2017","unstructured":"Nhut-Minh Ho and Weng-Fai Wong . 2017 . Exploiting half precision arithmetic in Nvidia GPUs . In Proceedings of the 2017 IEEE High Performance Extreme Computing Conference (HPEC). IEEE, 1--7. Nhut-Minh Ho and Weng-Fai Wong. 2017. Exploiting half precision arithmetic in Nvidia GPUs. In Proceedings of the 2017 IEEE High Performance Extreme Computing Conference (HPEC). IEEE, 1--7."},{"key":"e_1_2_1_62_1","volume-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications. CoRR abs\/1704.04861","author":"Howard Andrew G.","year":"2017","unstructured":"Andrew G. Howard , Menglong Zhu , Bo Chen , Dmitry Kalenichenko , Weijun Wang , Tobias Weyand , Marco Andreetto , and Hartwig Adam . 2017. MobileNets: Efficient convolutional neural networks for mobile vision applications. CoRR abs\/1704.04861 ( 2017 ). Andrew G. Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, and Hartwig Adam. 2017. MobileNets: Efficient convolutional neural networks for mobile vision applications. CoRR abs\/1704.04861 (2017)."},{"key":"e_1_2_1_63_1","unstructured":"Brody Huval Tao Wang Sameep Tandon Jeff Kiske Will Song Joel Pazhayampallil Mykhaylo Andriluka Pranav Rajpurkar Toki Migimatsu Royce Cheng-Yue etal 2015. An empirical evaluation of deep learning on highway driving. arXiv preprint arXiv:1504.01716 (2015).  Brody Huval Tao Wang Sameep Tandon Jeff Kiske Will Song Joel Pazhayampallil Mykhaylo Andriluka Pranav Rajpurkar Toki Migimatsu Royce Cheng-Yue et al. 2015. An empirical evaluation of deep learning on highway driving. arXiv preprint arXiv:1504.01716 (2015)."},{"key":"e_1_2_1_64_1","volume-title":"Proceedings of the 32nd International Conference on Machine Learning\u2014Volume 37 (ICML\u201915)","author":"Ioffe Sergey","year":"2015","unstructured":"Sergey Ioffe and Christian Szegedy . 2015 . Batch normalization: Accelerating deep network training by reducing internal covariate shift . In Proceedings of the 32nd International Conference on Machine Learning\u2014Volume 37 (ICML\u201915) . JMLR.org, 448--456. http:\/\/dl.acm.org\/citation.cfm?id&equals;3045118.3045167. Sergey Ioffe and Christian Szegedy. 2015. Batch normalization: Accelerating deep network training by reducing internal covariate shift. In Proceedings of the 32nd International Conference on Machine Learning\u2014Volume 37 (ICML\u201915). JMLR.org, 448--456. http:\/\/dl.acm.org\/citation.cfm?id&equals;3045118.3045167."},{"key":"e_1_2_1_65_1","volume-title":"Compression of deep convolutional neural networks for fast and low power mobile applications. arXiv preprint arXiv:1511.06530","author":"Kim Yong-Deok","year":"2015","unstructured":"Yong-Deok Kim , Eunhyeok Park , Sungjoo Yoo , Taelim Choi , Lu Yang , and Dongjun Shin . 2015. Compression of deep convolutional neural networks for fast and low power mobile applications. arXiv preprint arXiv:1511.06530 ( 2015 ). Yong-Deok Kim, Eunhyeok Park, Sungjoo Yoo, Taelim Choi, Lu Yang, and Dongjun Shin. 2015. Compression of deep convolutional neural networks for fast and low power mobile applications. arXiv preprint arXiv:1511.06530 (2015)."},{"key":"e_1_2_1_66_1","volume-title":"Proceedings of the 3rd International Conference on Learning Representations (ICLR\u201915)","author":"Diederik","unstructured":"Diederik P. Kingma and Jimmy Ba. 2015. Adam: A method for stochastic optimization . In Proceedings of the 3rd International Conference on Learning Representations (ICLR\u201915) . Diederik P. Kingma and Jimmy Ba. 2015. Adam: A method for stochastic optimization. In Proceedings of the 3rd International Conference on Learning Representations (ICLR\u201915)."},{"key":"e_1_2_1_67_1","volume-title":"The IEEE Conference on Computer Vision and Pattern Recognition (CVPR).","author":"Kreiss Sven","year":"2019","unstructured":"Sven Kreiss , Lorenzo Bertoni , and Alexandre Alahi . 2019 . PifPaf: Composite fields for human pose estimation . In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Sven Kreiss, Lorenzo Bertoni, and Alexandre Alahi. 2019. PifPaf: Composite fields for human pose estimation. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)."},{"key":"e_1_2_1_68_1","volume-title":"Deep reinforcement learning: An overview. arXiv preprint arXiv:1701.07274","author":"Yuxi Li.","year":"2017","unstructured":"Yuxi Li. 2017. Deep reinforcement learning: An overview. arXiv preprint arXiv:1701.07274 ( 2017 ). Yuxi Li. 2017. Deep reinforcement learning: An overview. arXiv preprint arXiv:1701.07274 (2017)."},{"key":"e_1_2_1_69_1","volume-title":"Efficient winograd or cook-toom convolution kernel implementation on widely used mobile CPUs. arXiv preprint arXiv:1903.01521","author":"Maji Partha","year":"2019","unstructured":"Partha Maji , Andrew Mundy , Ganesh Dasika , Jesse Beu , Matthew Mattina , and Robert Mullins . 2019. Efficient winograd or cook-toom convolution kernel implementation on widely used mobile CPUs. arXiv preprint arXiv:1903.01521 ( 2019 ). Partha Maji, Andrew Mundy, Ganesh Dasika, Jesse Beu, Matthew Mattina, and Robert Mullins. 2019. Efficient winograd or cook-toom convolution kernel implementation on widely used mobile CPUs. arXiv preprint arXiv:1903.01521 (2019)."},{"key":"e_1_2_1_70_1","doi-asserted-by":"publisher","DOI":"10.1080\/00220485.1991.10844705"},{"key":"e_1_2_1_71_1","volume-title":"Proceedings of the 14th Python in Science Conference, K. Huff and J. Bergstra (Eds.). 18--25","author":"McFee B.","unstructured":"B. McFee , C. Raffel , D. Liang , D. P.W. Ellis , M. McVicar , E. Battenberg , and O. Nieto . 2015. librosa: Audio and music signal analysis in Python . In Proceedings of the 14th Python in Science Conference, K. Huff and J. Bergstra (Eds.). 18--25 . B. McFee, C. Raffel, D. Liang, D. P.W. Ellis, M. McVicar, E. Battenberg, and O. Nieto. 2015. librosa: Audio and music signal analysis in Python. In Proceedings of the 14th Python in Science Conference, K. Huff and J. Bergstra (Eds.). 18--25."},{"key":"e_1_2_1_72_1","volume-title":"Proceedings of the 27th International Conference on Machine Learning (ICML\u201910)","author":"Nair Vinod","unstructured":"Vinod Nair and Geoffrey E. Hinton . 2010. Rectified linear units improve restricted boltzmann machines . In Proceedings of the 27th International Conference on Machine Learning (ICML\u201910) . Omnipress, 807--814. http:\/\/dl.acm.org\/citation.cfm?id&equals;3104322.3104425. Vinod Nair and Geoffrey E. Hinton. 2010. Rectified linear units improve restricted boltzmann machines. In Proceedings of the 27th International Conference on Machine Learning (ICML\u201910). Omnipress, 807--814. http:\/\/dl.acm.org\/citation.cfm?id&equals;3104322.3104425."},{"key":"e_1_2_1_73_1","volume-title":"Proceedings of the 2018 Design, Automation 8 Test in Europe Conference 8 Exhibition (DATE). IEEE, 827--832","author":"Shafique Muhammad","year":"2018","unstructured":"Muhammad Shafique , Theocharis Theocharides , Christos-Savvas Bouganis , Muhammad Abdullah Hanif , Faiq Khalid , Rehan Haf\u0131z , and Semeen Rehman . 2018 . An overview of next-generation architectures for machine learning: Roadmap, opportunities and challenges in the IoT era . In Proceedings of the 2018 Design, Automation 8 Test in Europe Conference 8 Exhibition (DATE). IEEE, 827--832 . Muhammad Shafique, Theocharis Theocharides, Christos-Savvas Bouganis, Muhammad Abdullah Hanif, Faiq Khalid, Rehan Haf\u0131z, and Semeen Rehman. 2018. An overview of next-generation architectures for machine learning: Roadmap, opportunities and challenges in the IoT era. In Proceedings of the 2018 Design, Automation 8 Test in Europe Conference 8 Exhibition (DATE). IEEE, 827--832."},{"key":"e_1_2_1_74_1","doi-asserted-by":"publisher","DOI":"10.1109\/TSA.2002.800560"},{"key":"e_1_2_1_75_1","volume-title":"Proceedings of the 2019 IEEE 39th International Conference on Distributed Computing Systems (ICDCS). 1630--1640","author":"Yao S.","year":"2019","unstructured":"S. Yao , Y. Hao , Y. Zhao , A. Piao , H. Shao , D. Liu , S. Liu , S. Hu , D. Weerakoon , K. Jayarajah , A. Misra , and T. Abdelzaher . 2019. Eugene: Towards deep intelligence as a service . In Proceedings of the 2019 IEEE 39th International Conference on Distributed Computing Systems (ICDCS). 1630--1640 . DOI:https:\/\/doi.org\/10.1109\/ICDCS. 2019 .00162 10.1109\/ICDCS.2019.00162 S. Yao, Y. Hao, Y. Zhao, A. Piao, H. Shao, D. Liu, S. Liu, S. Hu, D. Weerakoon, K. Jayarajah, A. Misra, and T. Abdelzaher. 2019. Eugene: Towards deep intelligence as a service. In Proceedings of the 2019 IEEE 39th International Conference on Distributed Computing Systems (ICDCS). 1630--1640. DOI:https:\/\/doi.org\/10.1109\/ICDCS.2019.00162"},{"key":"e_1_2_1_76_1","unstructured":"Y. Zhang N. Suda L. Lai and V. Chandra. 2018. Hello edge: Keyword spotting on microcontrollers. ArXiv e-prints (Feb. 2018). arxiv:cs.SD\/1711.07128.  Y. Zhang N. Suda L. Lai and V. Chandra. 2018. Hello edge: Keyword spotting on microcontrollers. ArXiv e-prints (Feb. 2018). arxiv:cs.SD\/1711.07128."}],"container-title":["ACM Transactions on Internet of Things"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3403572","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3403572","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T21:31:46Z","timestamp":1750195906000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3403572"}},"subtitle":["End-to-end integration of data, algorithms, and deployment tools"],"short-title":[],"issued":{"date-parts":[[2020,8,4]]},"references-count":76,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2020,11,30]]}},"alternative-id":["10.1145\/3403572"],"URL":"https:\/\/doi.org\/10.1145\/3403572","relation":{},"ISSN":["2691-1914","2577-6207"],"issn-type":[{"value":"2691-1914","type":"print"},{"value":"2577-6207","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,8,4]]},"assertion":[{"value":"2019-06-01","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2020-05-01","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2020-08-04","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}