Abstract
Resource-constrained IoT devices, such as sensors and actuators, have become ubiquitous in recent years. This has led to the generation of large quantities of data in real-time, which is an appealing target for AI systems. However, deploying machine learning models on such end-devices is nearly impossible. A typical solution involves offloading data to external computing systems (such as cloud servers) for further processing but this worsens latency, leads to increased communication costs, and adds to privacy concerns. To address this issue, efforts have been made to place additional computing devices at the edge of the network, i.e., close to the IoT devices where the data is generated. Deploying machine learning systems on such edge computing devices alleviates the above issues by allowing computations to be performed close to the data sources. This survey describes major research efforts where machine learning systems have been deployed at the edge of computer networks, focusing on the operational aspects including compression techniques, tools, frameworks, and hardware used in successful applications of intelligent edge systems.
- Amina Adadi and Mohammed Berrada. 2018. Peeking inside the black-box: A survey on explainable artificial intelligence (XAI). IEEE Access 6 (2018), 52138–52160.Google ScholarCross Ref
- Adafruit. 2019. Micro Speech Demo. Retrieved from https://learn.adafruit.com/tensorflow-lite-for-edgebadge-kit-quickstart/micro-speech-demo.Google Scholar
- M. Ali, A. Anjum, M. U. Yaseen, A. R. Zamani, D. Balouek-Thomert, O. Rana, and M. Parashar. 2018. Edge enhanced deep learning system for large-scale video stream analytics. In IEEE 2nd International Conference on Fog and Edge Computing (ICFEC). 1–10. DOI:https://doi.org/10.1109/CFEC.2018.8358733Google Scholar
- Alasdair Allan. 2018. Deep Learning at the Edge on an Arm Cortex-powered Camera Board. Retrieved from https://blog.hackster.io/deep-learning-at-the-edge-on-an-arm-cortex-powered-camera-board-3ca16eb60ef7.Google Scholar
- Alasdair Allan. 2019. Benchmarking Edge Computing. Retrieved from https://medium.com/@aallan/benchmarking-edge-computing-ce3f13942245.Google Scholar
- Alasdair Allan. 2019. Benchmarking the Xnor AI2GO Platform on the Raspberry Pi. Retrieved from https://blog.hackster.io/benchmarking-the-xnor-ai2go-platform-on-the-raspberry-pi-628a82af8aea.Google Scholar
- Alasdair Allan. 2019. Hands-on with the SmartEdge Agile. Retrieved from https://blog.hackster.io/hands-on-with-the-smartedge-agile-b7b7f02b5d4b.Google Scholar
- Alasdair Allan. 2019. Measuring Machine Learning. Retrieved from https://towardsdatascience.com/measuring-machine-learning-945a47bd3750.Google Scholar
- G. Ananthanarayanan, P. Bahl, P. Bodík, K. Chintalapudi, M. Philipose, L. Ravindranath, and S. Sinha. 2017. Real-time video analytics: The killer app for edge computing. Computer 50, 10 (2017), 58–67.Google ScholarDigital Library
- Ganesh Ananthanarayanan, Victor Bahl, Landon Cox, Alex Crown, Shadi Nogbahi, and Yuanchao Shu. 2019. Video analytics—Killer app for edge computing. In 17th International Conference on Mobile Systems, Applications, and Services (MobiSys’19). ACM, New York, NY, 695–696. DOI:https://doi.org/10.1145/3307334.3328589 Google ScholarDigital Library
- Andrej Karpathy. 2019. PyTorch at Tesla. Retrieved from https://www.youtube.com/watch?v=oBklltKXtDE.Google Scholar
- ARM Limited. Machine Learning ARM ML Processor. Retrieved on July 25, 2021 from https://developer.arm.com/ip-products/processors/machine-learning.Google Scholar
- Asha Barbaschow. 2018. VMware looking towards IoT and the edge. Retrieved from https://www.zdnet.com/article/vmware-looking-towards-iot-and-the-edge/.Google Scholar
- M. Barnell, C. Raymond, C. Capraro, D. Isereau, C. Cicotta, and N. Stokes. 2018. High-performance computing (HPC) and machine learning demonstrated in flight using Agile Condor. In IEEE High Performance Extreme Computing Conference (HPEC). 1–4.Google Scholar
- B. Barry, C. Brick, F. Connor, D. Donohoe, D. Moloney, R. Richmond, M. O’Riordan, and V. Toma. 2015. Always-on vision processing unit for mobile applications. IEEE Micro 35, 2 (Mar. 2015), 56–66. DOI:https://doi.org/10.1109/MM.2015.10Google ScholarDigital Library
- Sourav Bhattacharya and Nicholas D. Lane. 2016. Sparsification and separation of deep learning layers for constrained resource inference on wearables. In 14th ACM Conference on Embedded Network Sensor Systems CD-ROM (SenSys’16). ACM, New York, NY, 176–189. Google ScholarDigital Library
- Sumon Biswas and Hridesh Rajan. 2020. Do the machine learning models on a crowd sourced platform exhibit bias? An empirical study on model fairness. In 28th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering. Google ScholarDigital Library
- Keith Bonawitz, Vladimir Ivanov, Ben Kreuter, Antonio Marcedone, H. Brendan McMahan, Sarvar Patel, Daniel Ramage, Aaron Segal, and Karn Seth. 2017. Practical secure aggregation for privacy-preserving machine learning. In ACM SIGSAC Conference on Computer and Communications Security. 1175–1191. Google ScholarDigital Library
- Andrew A. Borkowski, Catherine P. Wilson, Steven A. Borkowski, Lauren A. Deland, and Stephen M. Mastorides. 2019. Using Apple machine learning algorithms to detect and subclassify non-small cell lung cancer. Arxiv E-prints 1808.08230 (January 2019).Google Scholar
- Brandon Butler. 2017. What is edge computing and how it’s changing the network. Network World (2017). DOI:https://www.networkworld.com/article/3224893/what-is-edge-computing-and-how-it-s-changing-the-network.html.Google Scholar
- S. Cass. 2019. Taking AI to the edge: Google’s TPU now comes in a maker-friendly package. IEEE Spectrum 56, 5 (May 2019), 16–17. DOI:https://doi.org/10.1109/MSPEC.2019.8701189Google ScholarCross Ref
- W. Chang, L. Chen, and K. Su. 2019. DeepCrash: A deep learning-based internet of vehicles system for head-on and single-vehicle accident detection with emergency notification. IEEE Access 7 (2019), 148163–148175.Google ScholarCross Ref
- X. Chang, W. Li, C. Xia, J. Ma, J. Cao, S. U. Khan, and A. Y. Zomaya. 2018. From insight to impact: Building a sustainable edge computing platform for smart homes. In IEEE 24th International Conference on Parallel and Distributed Systems (ICPADS). 928–936. DOI:https://doi.org/10.1109/PADSW.2018.8644647Google Scholar
- G. Chen, C. Parada, and G. Heigold. 2014. Small-footprint keyword spotting using deep neural networks. In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). 4087–4091. DOI:https://doi.org/10.1109/ICASSP.2014.6854370Google Scholar
- J. Chen and X. Ran. 2019. Deep learning with edge computing: A review. Proc. IEEE 107, 8 (Aug. 2019), 1655–1674.Google ScholarCross Ref
- Min Chen, Yuanwen Tian, Giancarlo Fortino, Jing Zhang, and Iztok Humar. 2018. Cognitive internet of vehicles. Comput. Commun. 120 (2018), 58–70. Google ScholarDigital Library
- Tianqi Chen, Thierry Moreau, Ziheng Jiang, Lianmin Zheng, Eddie Yan, Haichen Shen, Meghan Cowan, Leyuan Wang, Yuwei Hu, Luis Ceze, Carlos Guestrin, and Arvind Krishnamurthy. 2018. TVM: An automated end-to-end optimizing compiler for deep learning. In 13th USENIX Symposium on Operating Systems Design and Implementation (OSDI’18). USENIX Association, 578–594. https://www.usenix.org/conference/osdi18/presentation/chen. Google ScholarDigital Library
- Y. Chen, A. Wu, M. A. Bayoumi, and F. Koushanfar. 2013. Editorial low-power, intelligent, and secure solutions for realization of internet of things. IEEE J. Emerg. Select. Topics Circ. Syst. 3, 1 (Mar. 2013), 1–4. DOI:https://doi.org/10.1109/JETCAS.2013.2244771Google ScholarCross Ref
- Yu Cheng, Duo Wang, Pan Zhou, and Tao Zhang. 2017. A survey of model compression and acceleration for deep neural networks. Arxiv E-prints 1710.09282 (2017).Google Scholar
- Sandeep Chinchali, Apoorva Sharma, James Harrison, Amine Elhafsi, Daniel Kang, Evgenya Pergament, Eyal Cidon, Sachin Katti, and Marco Pavone. 2021. Network offloading policies for cloud robotics: a learning-based approach. Autonomous Robots (2021), 1–16. https://doi.org/10.1007/s10514-021-09987-4Google Scholar
- Tejalal Choudhary, Vipul Mishra, Anurag Goswami, and Jagannathan Sarangapani. 2020. A comprehensive survey 1109 on model compression and acceleration. Artif. Intell. Rev. (2020), 1–43.Google Scholar
- Christine Long. 2019. BeagleBone AI Makes a Sneak Preview. Retrieved from https://beagleboard.org/blog/2019-05-16-beaglebone-ai-preview.Google Scholar
- N. Curukogle and B. M. Ozyildirim. 2018. Deep learning on mobile systems. In Innovations in Intelligent Systems and Applications Conference (ASYU). 1–4. DOI:https://doi.org/10.1109/ASYU.2018.8554039Google Scholar
- A. Das, M. Degeling, X. Wang, J. Wang, N. Sadeh, and M. Satyanarayanan. 2017. Assisting users in a world full of cameras: A privacy-aware infrastructure for computer vision applications. In IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). 1387–1396. DOI:https://doi.org/10.1109/CVPRW.2017.181Google Scholar
- Rustem Dautov, Salvatore Distefano, Dario Bruneo, Francesco Longo, Giovanni Merlino, Antonio Puliafito, and Rajkumar Buyya. 2018. Metropolitan intelligent surveillance systems for urban areas by harnessing IoT and edge computing paradigms. Softw., Pract. Exper. 48 (2018), 1475–1492.Google ScholarCross Ref
- G. Ditzler, M. Roveri, C. Alippi, and R. Polikar. 2015. Learning in nonstationary environments: A survey. IEEE Comput. Intell. Mag. 10, 4 (2015), 12–25.Google ScholarDigital Library
- Utsav Drolia, Katherine Guo, and Priya Narasimhan. 2017. Precog: Prefetching for image recognition applications at the edge. In 2nd ACM/IEEE Symposium on Edge Computing (SEC’17). ACM, New York, NY. DOI:https://doi.org/10.1145/3132211.3134456 Google ScholarDigital Library
- SparkFun Electronics. SparkFun Edge Hookup Guide. Retreived July 25, 2021 from https://learn.sparkfun.com/tutorials/sparkfun-edge-hookup-guide/all.Google Scholar
- A. R. Elias, N. Golubovic, C. Krintz, and R. Wolski. 2017. Where’s the bear?—Automating wildlife image processing using IoT and edge cloud systems. In IEEE/ACM 2nd International Conference on Internet-of-Things Design and Implementation (IoTDI). 247–258. Google ScholarDigital Library
- E. Ezra Tsur, E. Madar, and N. Danan. 2018. Code generation of graph-based vision processing for multiple CUDA cores SoC Jetson TX. In IEEE 12th International Symposium on Embedded Multicore/Many-core Systems-on-Chip (MCSoC). 1–7. DOI:https://doi.org/10.1109/MCSoC2018.2018.00013Google ScholarCross Ref
- Z. Feng, S. George, J. Harkes, P. Pillai, R. Klatzky, and M. Satyanarayanan. 2018. Edge-based discovery of training data for machine learning. In IEEE/ACM Symposium on Edge Computing (SEC). 145–158.Google Scholar
- E. Flamand, D. Rossi, F. Conti, I. Loi, A. Pullini, F. Rotenberg, and L. Benini. 2018. GAP-8: A RISC-V SoC for AI at the edge of the IoT. In IEEE 29th International Conference on Application-specific Systems, Architectures and Processors (ASAP). 1–4. DOI:https://doi.org/10.1109/ASAP.2018.8445101Google Scholar
- David Floyer. 2015. The Vital Role of Edge Computing in the Internet of Things. Retrieved from https://wikibon.com/the-vital-role-of-edge-computing-in-the-internet-of-things.Google Scholar
- The Linux Foundation. Accessed:. The Open Platform for the IoT Edge. Retrieved on July 25, 2021 from https://www.edgexfoundry.org.Google Scholar
- Sorelle A. Friedler, Carlos Scheidegger, Suresh Venkatasubramanian, Sonam Choudhary, Evan P. Hamilton, and Derek Roth. 2019. A comparative study of fairness-enhancing interventions in machine learning. In Proceedings of the Conference on Fairness, Accountability, and Transparency (FAT*’19). Association for Computing Machinery, New York, NY, 329–338. Google ScholarDigital Library
- C. Gao, Antonio Rios-Navarro, Xi Chen, T. Delbrück, and Shih-Chii Liu. 2020. EdgeDRNN: Enabling low-latency recurrent neural network edge inference. In 2nd IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS). 41–45.Google ScholarCross Ref
- Robin C. Geyer, Tassilo Klein, and Moin Nabi. 2017. Differentially private federated learning: A client level perspective. CoRR abs/1712.07557 (2017).Google Scholar
- A. Ghoneim, G. Muhammad, S. U. Amin, and B. Gupta. 2018. Medical image forgery detection for smart healthcare. IEEE Commun. Mag. 56, 4 (Apr. 2018), 33–37. DOI:https://doi.org/10.1109/MCOM.2018.1700817Google ScholarCross Ref
- Sridhar Gopinath, Nikhil Ghanathe, Vivek Seshadri, and Rahul Sharma. 2019. Compiling KB-sized machine learning models to tiny IoT devices. In 40th ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI’19). ACM, New York, NY, 79–95. DOI:https://doi.org/10.1145/3314221.3314597 Google ScholarDigital Library
- Chirag Gupta, Arun Sai Suggala, Ankit Goyal, Harsha Vardhan Simhadri, Bhargavi Paranjape, Ashish Kumar, Saurabh Goyal, Raghavendra Udupa, Manik Varma, and Prateek Jain. 2017. ProtoNN: Compressed and accurate kNN for resource-scarce devices. In Proceedings of the 34th International Conference on Machine Learning (Proceedings of Machine Learning Research), Vol. 70. PMLR, 1331–1340. Retrieved from http://proceedings.mlr.press/v70/gupta17a.html. Google ScholarDigital Library
- Suyog Gupta, Ankur Agrawal, Kailash Gopalakrishnan, and Pritish Narayanan. 2015. Deep learning with limited numerical precision. In 32nd International Conference on International Conference on Machine Learning - Volume 37 (ICML’15). JMLR.org, 1737–1746. Retrieved from http://dl.acm.org/citation.cfm?id=3045118.3045303. Google ScholarDigital Library
- Ramyad Hadidi, Jiashen Cao, Michael S. Ryoo, and Hyesoon Kim. 2019. Robustly executing DNNs in IoT systems using coded distributed computing. In 56th Design Automation Conference 2019 (DAC’19). Association for Computing Machinery, New York, NY. Google ScholarDigital Library
- R. Hadidi, J. Cao, M. Woodward, M. S. Ryoo, and H. Kim. 2018. Distributed perception by collaborative robots. IEEE Robot. Autom. Lett. 3, 4 (2018), 3709–3716.Google ScholarCross Ref
- Song Han, Huizi Mao, and William J. Dally. 2015. Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding. arXiv:https://arxiv.org/abs/1510.00149.Google Scholar
- Awni Hannun, Carl Case, Jared Casper, Bryan Catanzaro, Greg Diamos, Erich Elsen, Ryan Prenger, Sanjeev Satheesh, Shubho Sengupta, Adam Coates, and Andrew Y. Ng. 2014. Deep Speech: Scaling up end-to-end speech recognition. arXiv:https://arxiv.org/abs/1412.5567.Google Scholar
- Richard Harper. 2003. Inside the Smart House. Springer-Verlag, Berlin. Google ScholarDigital Library
- Evan Hennis, Mark Deoust, and Billy Lamberta. 2019. TensorFlow Lite Speech Command Recognition Android Demo. Retrieved from https://github.com/tensorflow/examples/tree/master/lite/examples/speech_commands/android.Google Scholar
- Jacob Hochstetler, Rahul Padidela, Qing Chen, Qiang Yang, and Songnian Fu. 2018. Embedded deep learning for vehicular edge computing. In IEEE/ACM Symposium on Edge Computing (SEC). 341–343.Google ScholarCross Ref
- 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. arXiv:https://arxiv.org/abs/1704.04861.Google Scholar
- C. C.-H. Hsu, M. Y.-C. Wang, H. C. H. Shen, R. H. Chiang, and C. H. P. Wen. 2017. FallCare+: An IoT surveillance system for fall detection. In International Conference on Applied System Innovation (ICASI). 921–922.Google ScholarCross Ref
- Z. Huai, B. Ding, H. Wang, M. Geng, and L. Zhang. 2019. Towards deep learning on resource-constrained robots: A crowdsourcing approach with model partition. In IEEE SmartWorld, Ubiquitous Intelligence Computing, Advanced Trusted Computing, Scalable Computing Communications, Cloud Big Data Computing, Internet of People and Smart City Innovation (SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI). 989–994.Google Scholar
- Itay Hubara, Matthieu Courbariaux, Daniel Soudry, Ran El-Yaniv, and Yoshua Bengio. 2016. Binarized neural networks. In 30th International Conference on Neural Information Processing Systems. 4114–4122. Google ScholarDigital Library
- Forrest N. Iandola, Song Han, Matthew W. Moskewicz, Khalid Ashraf, William J. Dally, and Kurt Keutzer. 2016. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5 MB model size. arXiv:https://arxiv.org/abs/1602.07360.Google Scholar
- Michaela Iorga, Larry B. Feldman, Robert Barton, Michael Martin, Nedim S. Goren, and Charif Mahmoudi. 2018. Fog Computing Conceptual Model. DOI:https://doi.org/10.6028/NIST.SP.500-325Google Scholar
- D. Isereau, C. Capraro, E. Cote, M. Barnell, and C. Raymond. 2017. Utilizing high-performance embedded computing, Agile Condor, for intelligent processing: An artificial intelligence platform for remotely piloted aircraft. In Intelligent Systems Conference (IntelliSys). 1155–1159. DOI:https://doi.org/10.1109/IntelliSys.2017.8324277Google Scholar
- Kaya Ismail. 2018. Edge Computing vs. Fog Computing: What’s the Difference?Retrieved from https://www.cmswire.com/information-management/edge-computing-vs-fog-computing-whats-the-difference/.Google Scholar
- R. Colin Johnson. 2019. Neural Learning on the Edge. Retrieved from https://cacm.acm.org/news/234063-neural-learning-on-the-edge/fulltext.Google Scholar
- Vinu Joseph, Ganesh L. Gopalakrishnan, Saurav Muralidharan, Michael Garland, and Animesh Garg. 2020. A programmable approach to neural network compression. IEEE Micro 40, 5 (Sep. 2020), 17–25.Google ScholarDigital Library
- Daniel Kang, Peter Bailis, and Matei Zaharia. 2019. Challenges and opportunities in DNN-based video analytics: A demonstration of the BlazeIt video query engine. In 9th Biennial Conference on Innovative Data Systems Research. Retrieved from http://cidrdb.org/cidr2019/papers/p141-kang-cidr19.pdf.Google Scholar
- Duseok Kang, Euiseok Kim, Inpyo Bae, Bernhard Egger, and Soonhoi Ha. 2018. C-GOOD: C-code generation framework for optimized on-device deep learning. In International Conference on Computer-Aided Design (ICCAD’18). ACM, New York, NY. DOI:https://doi.org/10.1145/3240765.3240786 Google ScholarDigital Library
- Gorkem Kar, Shubham Jain, Marco Gruteser, Fan Bai, and Ramesh Govindan. 2017. Real-time traffic estimation at vehicular edge nodes. In 2nd ACM/IEEE Symposium on Edge Computing (SEC’17). ACM, New York, NY. DOI:https://doi.org/10.1145/3132211.3134461 Google ScholarDigital Library
- Jae-Yun Kim and Soo-Mook Moon. 2018. Blockchain-based edge computing for deep neural network applications. In Workshop on INTelligent Embedded Systems Architectures and Applications (INTESA’18). Association for Computing Machinery, New York, NY, 53–55. Google ScholarDigital Library
- Ashish Kumar, Saurabh Goyal, and Manik Varma. 2017. Resource-efficient machine learning in 2 KB RAM for the internet of things. In 34th International Conference on Machine Learning (Proceedings of Machine Learning Research), Vol. 70. PMLR, 1935–1944. Retrieved from http://proceedings.mlr.press/v70/kumar17a.html. Google ScholarDigital Library
- Aditya Kusupati, Don Dennis, Chirag Gupta, Ashish Kumar, Shishir Patil, and Harsha Simhadri. 2021. The EdgeML Library: An ML library for machine learning on the Edge. Retrieved on July 25, 2021 from https://github.com/Microsoft/EdgeML.Google Scholar
- Aditya Kusupati, Manish Singh, Kush Bhatia, Ashish Kumar, Prateek Jain, and Manik Varma. 2018. FastGRNN: A fast, accurate, stable and tiny kilobyte sized gated recurrent neural network. In Advances in Neural Information Processing Systems 31. Curran Associates, Inc., 9017–9028. Retrieved from http://papers.nips.cc/paper/8116-fastgrnn-a-fast-accurate-stable-and-tiny-kilobyte-sized-gated-recurrent-neural-network.pdf. Google ScholarDigital Library
- Gant Laborde. 2019. Perf Machine Learning on Rasp Pi. Retrieved from https://medium.com/free-code-camp/perf-machine-learning-on-rasp-pi-51101d03dba2.Google Scholar
- Liangzhen Lai and Naveen Suda. 2018. Enabling deep learning at the IoT edge. In International Conference on Computer-Aided Design (ICCAD’18). ACM, New York, NY. DOI:https://doi.org/10.1145/3240765.3243473 Google ScholarDigital Library
- N. D. Lane, S. Bhattacharya, P. Georgiev, C. Forlivesi, L. Jiao, L. Qendro, and F. Kawsar. 2016. DeepX: A software accelerator for low-power deep learning inference on mobile devices. In 15th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN). 1–12. DOI:https://doi.org/10.1109/IPSN.2016.7460664 Google ScholarDigital Library
- S. Lee, K. Son, H. Kim, and J. Park. 2017. Car plate recognition based on CNN using embedded system with GPU. In 10th International Conference on Human System Interactions (HSI). 239–241. DOI:https://doi.org/10.1109/HSI.2017.8005037Google Scholar
- D. Li, T. Salonidis, N. V. Desai, and M. C. Chuah. 2016. DeepCham: Collaborative edge-mediated adaptive deep learning for mobile object recognition. In IEEE/ACM Symposium on Edge Computing (SEC). 64–76. DOI:https://doi.org/10.1109/SEC.2016.38Google ScholarCross Ref
- En Li, Zhi Zhou, and Xu Chen. 2018. Edge intelligence: On-demand deep learning model co-inference with device-edge synergy. In Workshop on Mobile Edge Communications (MECOMM’18). ACM, New York, NY, 31–36. Google ScholarDigital Library
- Mingzhen Li, Yi Liu, Xiaoyan Liu, Qingxiao Sun, Xin You, Hailong Yang, Zhongzhi Luan, Lin Gan, Guangwen Yang, and Depei Qian. 2020. The Deep Learning Compiler: A Comprehensive Survey. arXiv:https://arxiv.org/abs/2002.03794.Google Scholar
- Tian Li, Anit Kumar Sahu, Ameet Talwalkar, and Virginia Smith. 2019. Federated learning: Challenges, methods, and future directions. Arxiv Eprints abs/1908.07873 (2019).Google Scholar
- Le Liang, Hao Ye, and Geoffrey Ye Li. 2018. Toward intelligent vehicular networks: A machine learning framework. IEEE Internet Things J. 6, 1 (2018), 124–135.Google ScholarCross Ref
- Wei Yang Bryan Lim, Nguyen Cong Luong, Dinh Thai Hoang, Yutao Jiao, Ying-Chang Liang, Qiang Yang, Dusit Niyato, and Chunyan Miao. 2020. Federated learning in mobile edge networks: A comprehensive survey. IEEE Commun. Surv. Tutor. 22, 3 (2020), 2031–2063.Google ScholarCross Ref
- Yujun Lin, Song Han, Huizi Mao, Yu Wang, and William J. Dally. 2020. Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Training. arXiv:https://arxiv.org/abs/1712.01887.Google Scholar
- Zhong Qiu Lin, Audrey G. Chung, and Alexander Wong. 2018. EdgeSpeechNets: Highly Efficient Deep Neural Networks for Speech Recognition on the Edge. arXiv:https://arxiv.org/abs/1810.08559.Google Scholar
- C. Liu, Y. Cao, Y. Luo, G. Chen, V. Vokkarane, M. Yunsheng, S. Chen, and P. Hou. 2018. A new deep learning-based food recognition system for dietary assessment on an edge computing service infrastructure. IEEE Trans. Serv.Comput. 11, 2 (Mar. 2018), 249–261. DOI:https://doi.org/10.1109/TSC.2017.2662008Google ScholarCross Ref
- Lumin Liu, Jun Zhang, S. H. Song, and Khaled B. Letaief. 2019. Client-Edge-Cloud Hierarchical Federated Learning. arXiv:https://arxiv.org/abs/1905.06641Google Scholar
- L. Liu, X. Zhang, M. Qiao, and W. Shi. 2018. SafeShareRide: Edge-based attack detection in ridesharing services. In IEEE/ACM Symposium on Edge Computing (SEC). 17–29. DOI:https://doi.org/10.1109/SEC.2018.00009Google Scholar
- Qiang Liu, Siqi Huang, and Tao Han. 2017. Fast and accurate object analysis at the edge for mobile augmented reality: Demo. In 2nd ACM/IEEE Symposium on Edge Computing (SEC’17). ACM, New York, NY. DOI:https://doi.org/10.1145/3132211.3132458 Google ScholarDigital Library
- Wang Luping, Wang Wei, and Li Bo. 2019. CMFL: Mitigating communication overhead for federated learning. https://doi.org/10.1109/ICDCS.2019.00099Google Scholar
- Salma Abdel Magid, Francesco Petrini, and Behnam Dezfouli. 2020. Image classification on IoT edge devices: Profiling and modeling. Clust. Comput. 23, 2 (2020), 1025–1043.Google ScholarDigital Library
- Mohammad Saeid Mahdavinejad, Mohammadreza Rezvan, Mohammadamin Barekatain, Peyman Adibi, Payam Barnaghi, and Amit P. Sheth. 2018. Machine learning for internet of things data analysis: A survey. Dig. Commun. Netw. 4, 3 (2018), 161–175. DOI:https://doi.org/10.1016/j.dcan.2017.10.002Google ScholarCross Ref
- James Manyika, Michael Chui, Peter Bisson, Jonathan Woetzel, Richard Dobbs, Jacques Bughin, and Dan Aharon. 2015. The Internet of Things: Mapping the Value Behind the Hype. Technical Report. McKinsey and Company.Google Scholar
- J. Mao, X. Chen, K. W. Nixon, C. Krieger, and Y. Chen. 2017. MoDNN: Local distributed mobile computing system for deep neural network. In Design, Automation Test in Europe Conference Exhibition (DATE). 1396–1401. DOI:https://doi.org/10.23919/DATE.2017.7927211 Google ScholarDigital Library
- Y. Mao, S. Yi, Q. Li, J. Feng, F. Xu, and S. Zhong. 2018. Learning from differentially private neural activations with edge computing. In IEEE/ACM Symposium on Edge Computing (SEC). 90–102. DOI:https://doi.org/10.1109/SEC.2018.00014Google Scholar
- C. Marantos, N. Karavalakis, V. Leon, V. Tsoutsouras, K. Pekmestzi, and D. Soudris. 2018. Efficient support vector machines implementation on Intel/Movidius Myriad 2. In 7th International Conference on Modern Circuits and Systems Technologies (MOCAST). 1–4. DOI:https://doi.org/10.1109/MOCAST.2018.8376630Google Scholar
- V. Mazzia, A. Khaliq, F. Salvetti, and M. Chiaberge. 2020. Real-time apple detection system using embedded systems with hardware accelerators: An edge AI application. In 2019 IEEE 39th International Conference on Distributed Computing Systems (ICDCS). IEEE, 954–964.Google Scholar
- Bradley McDanel, Surat Teerapittayanon, and H. T. Kung. 2017. Embedded Binarized Neural Networks. arXiv:https://arxiv.org/abs/1709.02260.Google Scholar
- Brendan McMahan, Eider Moore, Daniel Ramage, Seth Hampson, and Blaise Aguera y Arcas. 2017. Communication-efficient learning of deep networks from decentralized data. In 20th International Conference on Artificial Intelligence and Statistics (Proceedings of Machine Learning Research), Vol. 54. PMLR, 1273–1282. Retrieved from http://proceedings.mlr.press/v54/mcmahan17a.html.Google Scholar
- Microsoft. 2018. Embedded Learning Library. https://microsoft.github.io/ELL/.Google Scholar
- Microsoft. 2019. Project Brainwave. Retrieved from https://www.microsoft.com/en-us/research/project/project-brainwave/.Google Scholar
- S. A. Miraftabzadeh, P. Rad, K. R. Choo, and M. Jamshidi. 2018. A privacy-aware architecture at the edge for autonomous real-time identity reidentification in crowds. IEEE Internet Things J. 5, 4 (Aug. 2018), 2936–2946. DOI:https://doi.org/10.1109/JIOT.2017.2761801Google ScholarCross Ref
- M. G. S. Murshed, J. J. Carroll, N. Khan, and F. Hussain. 2020. Resource-aware on-device deep learning for supermarket hazard detection. In 19th IEEE International Conference on Machine Learning and Applications (ICMLA). 871–876. DOI:https://doi.org/10.1109/ICMLA51294.2020.00142Google Scholar
- Pedro Navarro Lorente, Carlos Fernandez, Raul Borraz, and Diego Alonso. 2016. A machine learning approach to pedestrian detection for autonomous vehicles using high-definition 3D range data. Sensors 17 (12 2016), 18. DOI:https://doi.org/10.3390/s17010018Google Scholar
- Jiquan Ngiam, Aditya Khosla, Mingyu Kim, Juhan Nam, Honglak Lee, and Andrew Y. Ng. 2011. Multimodal deep learning. In 28th International Conference on International Conference on Machine Learning (ICML’11). Omnipress, 689–696. Retrieved from http://dl.acm.org/citation.cfm?id=3104482.3104569. Google ScholarDigital Library
- Seyed Yahya Nikouei, Yu Chen, Sejun Song, Ronghua Xu, Baek-Young Choi, and Timothy R. Faughnan. 2018. Intelligent surveillance as an edge network service: from Harr-Cascade, SVM to a lightweight CNN. arxiv:1805.00331.Google Scholar
- Takayuki Nishio and Ryo Yonetani. 2018. Client selection for federated learning with heterogeneous resources in mobile edge. In IEEE International Conference on Communications (ICC). 1–7.Google Scholar
- Henry Friday Nweke, Ying Wah Teh, Mohammed Ali Al-garadi, and Uzoma Rita Alo. 2018. Deep learning algorithms for human activity recognition using mobile and wearable sensor networks: State of the art and research challenges. Exp. Syst. Applic. 105 (2018), 233–261. DOI:https://doi.org/10.1016/j.eswa.2018.03.056Google ScholarCross Ref
- Samuel S. Ogden and Tian Guo. 2018. MODI: Mobile deep inference made efficient by edge computing. In USENIX Workshop on Hot Topics in Edge Computing (HotEdge’18). USENIX Association. Retrieved from https://www.usenix.org/conference/hotedge18/presentation/ogden.Google Scholar
- S. A. Osia, A. S. Shamsabadi, A. Taheri, H. R. Rabiee, and H. Haddadi. 2018. Private and scalable personal data analytics using hybrid edge-to-cloud deep learning. Computer 51, 5 (May 2018), 42–49.Google ScholarCross Ref
- Anand Oswal. 2018. Time to Get Serious About Edge Computing. Retrieved from https://blogs.cisco.com/enterprise/time-to-get-serious-about-edge-computing.Google Scholar
- Angshuman Parashar, Minsoo Rhu, Anurag Mukkara, Antonio Puglielli, Rangharajan Venkatesan, Brucek Khailany, Joel Emer, Stephen W. Keckler, and William J. Dally. 2017. SCNN: An accelerator for compressed-sparse convolutional neural networks. SIGARCH Comput. Archit. News 45, 2 (June 2017), 27–40. DOI:https://doi.org/10.1145/3140659.3080254Google ScholarDigital Library
- Donghyun Park, Seulgi Kim, Yelin An, and Jae-Yoon Jung. 2018. LiReD: A light-weight real-time fault detection system for edge computing using LSTM recurrent neural networks. Sensors 18, 7 (2018). DOI:https://doi.org/10.3390/s18072110Google Scholar
- Eunhyeok Park, Dongyoung Kim, and Sungjoo Yoo. 2018. Energy-efficient neural network accelerator based on outlier-aware low-precision computation. In 45th International Symposium on Computer Architecture (ISCA’18). IEEE Press, 688–698. Google ScholarDigital Library
- J. Park, S. Samarakoon, M. Bennis, and M. Debbah. 2019. Wireless network intelligence at the edge. Proc. IEEE 107, 11 (2019), 2204–2239.Google ScholarCross Ref
- David Patterson and Andrew Waterman. 2017. The RISC-V Reader: An Open Architecture Atlas. Strawberry Canyon LLC. Google ScholarDigital Library
- Diego Peteiro-Barral and Bertha Guijarro-Berdiñas. 2013. A survey of methods for distributed machine learning. Prog. Artif. Intell. 2, 1 (01 Mar. 2013), 1–11.Google Scholar
- K. Pradeep, K. Kamalavasan, R. Natheesan, and A. Pasqual. 2018. EdgeNet: SqueezeNet like convolution neural network on embedded FPGA. In 25th IEEE International Conference on Electronics, Circuits and Systems (ICECS). 81–84. DOI:https://doi.org/10.1109/ICECS.2018.8617876Google Scholar
- MD Abdur Rahman, M. Shamim Hossain, George Loukas, Elham Hassanain, Syed Sadiqur Rahman, Mohammed F. Alhamid, and Mohsen Guizani. 2018. Blockchain-based mobile edge computing framework for secure therapy applications. IEEE Access 6 (2018), 72469–72478.Google ScholarCross Ref
- Joseph Redmon and Ali Farhadi. 2018. YOLOv3: An Incremental Improvement. arXiv:https://arxiv.org/abs/1804.02767.Google Scholar
- Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. 2016. “Why should I trust you?”: Explaining the predictions of any classifier. In 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD’16). Association for Computing Machinery, New York, NY, 1135–1144. DOI:https://doi.org/10.1145/2939672.2939778 Google ScholarDigital Library
- Nadav Rotem, Jordan Fix, Saleem Abdulrasool, Garret Catron, Summer Deng, Roman Dzhabarov, Nick Gibson, James Hegeman, Meghan Lele, Roman Levenstein, Jack Montgomery, Bert Maher, Satish Nadathur, Jakob Olesen, Jongsoo Park, Artem Rakhov, Misha Smelyanskiy, and Man Wang. 2019. Glow: Graph Lowering Compiler Techniques for Neural Networks. arXiv:https://arxiv.org/abs/1805.00907.Google Scholar
- K. Rungsuptaweekoon, V. Visoottiviseth, and R. Takano. 2017. Evaluating the power efficiency of deep learning inference on embedded GPU systems. In 2nd International Conference on Information Technology (INCIT). 1–5. DOI:https://doi.org/10.1109/INCIT.2017.8257866Google Scholar
- M. Satyanarayanan. 2017. The emergence of edge computing. Computer 50, 1 (Jan. 2017), 30–39. DOI:https://doi.org/10.1109/MC.2017.9 Google ScholarDigital Library
- M. Satyanarayanan and N. Davies. 2019. Augmenting cognition through edge computing. Computer 52, 7 (July 2019), 37–46.Google ScholarCross Ref
- Ragini Sharma, Saman Biookaghazadeh, and Ming Zhao. 2018. Are existing knowledge transfer techniques effective for deep learning on edge devices? In 27th International Symposium on High-performance Parallel and Distributed Computing (HPDC’18). ACM, New York, NY, 15–16. Google ScholarDigital Library
- Weisong Shi, Jie Cao, Quan Zhang, Youhuizi Li, and Lanyu Xu. 2016. Edge computing: Vision and challenges. IEEE Internet Things J. 3 (2016), 637–646.Google ScholarCross Ref
- W. Shi and S. Dustdar. 2016. The promise of edge computing. Computer 49, 5 (May 2016), 78–81. DOI:https://doi.org/10.1109/MC.2016.145Google ScholarDigital Library
- Z. Song, B. Fu, F. Wu, Z. Jiang, L. Jiang, N. Jing, and X. Liang. 2020. DRQ: Dynamic region-based quantization for deep neural network acceleration. In ACM/IEEE 47th International Symposium on Computer Architecture (ISCA). 1010–1021. Google ScholarDigital Library
- Flávio Souza, Diego de Las Casas, Vinícius Flores, SunBum Youn, Meeyoung Cha, Daniele Quercia, and Virgílio Almeida. 2015. Dawn of the Selfie Era: The Whos, Wheres, and Hows of Selfies on Instagram. arXiv:https://arxiv.org/abs/1510.05700. Google ScholarDigital Library
- IBM Research Editorial Staff. 2017. IBM scientists team with The Weather Company to bring edge computing to life. Retrieved from https://www.ibm.com/blogs/research/2017/02/bringing-edge-computing-to-life/.Google Scholar
- Rafael Stahl, Zhuoran Zhao, Daniel Mueller-Gritschneder, Andreas Gerstlauer, and Ulf Schlichtmann. 2019. Fully distributed deep learning inference on resource-constrained edge devices. In Embedded Computer Systems: Architectures, Modeling, and Simulation, Dionisios N. Pnevmatikatos, Maxime Pelcat, and Matthias Jung (Eds.). Springer International Publishing, Cham, 77–90.Google Scholar
- Mingxing Tan and Quoc V. Le. 2019. EfficientNet: Rethinking model scaling for convolutional neural networks. Arxiv Eprints 1905.11946 (2019).Google Scholar
- B. Tang, Z. Chen, G. Hefferman, S. Pei, T. Wei, H. He, and Q. Yang. 2017. Incorporating intelligence in fog computing for big data analysis in smart cities. IEEE Trans. Industr. Inform. 13, 5 (Oct. 2017), 2140–2150.Google Scholar
- Jiaxi Tang, Rakesh Shivanna, Zhe Zhao, Dong Lin, Anima Singh, Ed H. Chi, and Sagar Jain. 2020. Understanding and improving knowledge distillation. Arxiv Eprints 2002.03532 (2020).Google Scholar
- Zeyi Tao and Qun Li. 2018. eSGD: Communication efficient distributed deep learning on the edge. In USENIX Workshop on Hot Topics in Edge Computing (HotEdge’18). USENIX Association. https://www.usenix.org/conference/hotedge18/presentation/tao.Google Scholar
- GreenWaves Technologies. 2018. GAP8 - GreenWaves. Retrieved from https://en.wikichip.org/wiki/greenwaves/gap8.Google Scholar
- GreenWaves Technologies. 2019. TF2GAP8. Retrieved from https://github.com/GreenWaves-Technologies/tf2gap8.Google Scholar
- S. Teerapittayanon, B. McDanel, and H. T. Kung. 2017. Distributed deep neural networks over the cloud, the edge and end devices. In IEEE 37th International Conference on Distributed Computing Systems (ICDCS). 328–339. DOI:https://doi.org/10.1109/ICDCS.2017.226Google Scholar
- Tom Simonite. 2019. The best algorithms struggle to recognize black faces equally. Retrieved from https://www.wired.com/story/best-algorithms-struggle-recognize-% black-faces-equally/.Google Scholar
- Shreshth Tuli, Nipam Basumatary, and Rajkumar Buyya. 2019. EdgeLens: Deep Learning based Object Detection in Integrated IoT, Fog and Cloud Computing Environments. arXiv:https://arxiv.org/abs/1906.11056.Google Scholar
- S. Ullah and D. Kim. 2020. Benchmarking Jetson platform for 3D point-cloud and hyper-spectral image classification. In IEEE International Conference on Big Data and Smart Computing (BigComp). 477–482.Google Scholar
- Sahar Voghoei, Navid Hashemi Tonekaboni, Jason G. Wallace, and Hamid Reza Arabnia. 2018. Deep learning at the edge. In International Conference on Computational Science and Computational Intelligence (CSCI). 895–901.Google ScholarCross Ref
- C. Wang, L. Gong, Q. Yu, X. Li, Y. Xie, and X. Zhou. 2017. DLAU: A scalable deep learning accelerator unit on FPGA. IEEE Trans. Comput.-aided Des. Integ. Circ. Syst. 36, 3 (Mar. 2017), 513–517. Google ScholarDigital Library
- J. Wang, Z. Feng, Z. Chen, S. George, M. Bala, P. Pillai, S. Yang, and M. Satyanarayanan. 2018. Bandwidth-efficient live video analytics for drones via edge computing. In IEEE/ACM Symposium on Edge Computing (SEC). 159–173. DOI:https://doi.org/10.1109/SEC.2018.00019Google Scholar
- Shiqiang Wang, Tiffany Tuor, Theodoros Salonidis, Kin K. Leung, Christian Makaya, Ting He, and Kevin S. Chan. 2018. When edge meets learning: Adaptive control for resource-constrained distributed machine learning. In IEEE Conference on Computer Communications. 63–71. DOI:https://doi.org/10.1109/INFOCOM.2018.8486403Google Scholar
- Xiaofei Wang, Yiwen Han, Victor C. M. Leung, Dusit Niyato, Xueqiang Yan, and Xu Chen. 2020. Convergence of edge computing and deep learning: A comprehensive survey. IEEE Commun. Surv. Tutor. 22, 2 (2020), 869–904.Google ScholarCross Ref
- Sally Ward-Foxton. 2019. AI at the Very, Very Edge. Retrieved from https://www.eetimes.com/document.asp?doc_id=1334918.Google Scholar
- Pete Warden. 2018. Speech Commands: A Dataset for Limited-vocabulary Speech Recognition. arXiv:https://arxiv.org/abs/1804.03209.Google Scholar
- Matt Welsh. 2019. True AI on a Raspberry Pi, with no extra hardware. Retrieved from https://medium.com/@mdwdotla/true-ai-on-a-raspberry-pi-with-no-extra-hardware-dcdbff12d068.Google Scholar
- R. Xu, S. Y. Nikouei, Y. Chen, A. Polunchenko, S. Song, C. Deng, and T. R. Faughnan. 2018. Real-time human objects tracking for smart surveillance at the edge. In IEEE International Conference on Communications (ICC). 1–6. DOI:https://doi.org/10.1109/ICC.2018.8422970Google Scholar
- Zhuangdi Xu, Harshit Gupta, and Umakishore Ramachandran. 2018. STTR: A system for tracking all vehicles all the time at the edge of the network. In 12th ACM International Conference on Distributed and Event-based Systems (DEBS’18). ACM, New York, NY, 124–135. Google ScholarDigital Library
- Tzu-Hsien Yang, Hsiang-Yun Cheng, Chia-Lin Yang, I-Ching Tseng, Han-Wen Hu, Hung-Sheng Chang, and Hsiang-Pang Li. 2019. Sparse ReRAM engine: Joint exploration of activation and weight sparsity in compressed neural networks. In 46th International Symposium on Computer Architecture (ISCA’19). Association for Computing Machinery, New York, NY, 236–249. DOI:https://doi.org/10.1145/3307650.3322271 Google ScholarDigital Library
- Shuochao Yao, Yiran Zhao, Aston Zhang, Lu Su, and Tarek F. Abdelzaher. 2017. Compressing deep neural network structures for sensing systems with a compressor-critic framework. arxiv:1706.01215.Google Scholar
- Mahmut Taha Yazici, Shadi Basurra, and Mohamed Medhat Gaber. 2018. Edge machine learning: Enabling smart internet of things applications. Big Data Cog. Comput. 2, 3 (2018), 26.Google ScholarCross Ref
- Jiecao Yu, Andrew Lukefahr, David Palframan, Ganesh Dasika, Reetuparna Das, and Scott Mahlke. 2017. Scalpel: Customizing DNN pruning to the underlying hardware parallelism. In 44th International Symposium on Computer Architecture (ISCA’17). Association for Computing Machinery, New York, NY, 548–560. Google ScholarDigital Library
- Qunsong Zeng, Yuqing Du, Kin K. Leung, and Kaibin Huang. 2019. Energy-efficient Radio Resource Allocation for Federated Edge Learning. arXiv:https://arxiv.org/abs/1907.06040.Google Scholar
- Jiaqi Zhang, Xiangru Chen, Mingcong Song, and Tao Li. 2019. Eager pruning: Algorithm and architecture support for fast training of deep neural networks. In 46th International Symposium on Computer Architecture (ISCA’19). Association for Computing Machinery, New York, NY, 292–303. DOI:https://doi.org/10.1145/3307650.3322263 Google ScholarDigital Library
- Jianhao Zhang, Yingwei Pan, Ting Yao, He Zhao, and Tao Mei. 2019. daBNN: A super fast inference framework for binary neural networks on arm devices. In 27th ACM International Conference on Multimedia. 2272–2275. DOI:https://doi.org/10.1145/3343031.3350534 Google ScholarDigital Library
- Xingzhou Zhang, Yifan Wang, and Weisong Shi. 2018. pCAMP: Performance comparison of machine learning packages on the edges. In USENIX Workshop on Hot Topics in Edge Computing (HotEdge’18). USENIX Association, Boston, MA. Retrieved from https://www.usenix.org/conference/hotedge18/presentation/zhang.Google Scholar
- X. Zhang, X. Zhou, M. Lin, and J. Sun. 2018. ShuffleNet: An extremely efficient convolutional neural network for mobile devices. In IEEE/CVF Conference on Computer Vision and Pattern Recognition. 6848–6856. DOI:https://doi.org/10.1109/CVPR.2018.00716Google ScholarCross Ref
- Zhuoran Zhao, Kamyar Mirzazad Barijough, and Andreas Gerstlauer. 2018. DeepThings: Distributed adaptive deep learning inference on resource-constrained IoT edge clusters. IEEE Trans. Comput.-aided Des. Integ. Circ. Syst. PP (10 2018), 1–1. DOI:https://doi.org/10.1109/TCAD.2018.2858384Google Scholar
- Z. Zhou, X. Chen, E. Li, L. Zeng, K. Luo, and J. Zhang. 2019. Edge intelligence: Paving the last mile of artificial intelligence with edge computing. Proc. IEEE 107, 8 (Aug. 2019), 1738–1762.Google ScholarCross Ref
- Guangxu Zhu, Dongzhu Liu, Yuqing Du, Changsheng You, Jun Zhang, and Kaibin Huang. 2018. Towards an Intelligent Edge: Wireless Communication Meets Machine Learning. arXiv:https://arxiv.org/abs/1809.00343.Google Scholar
Index Terms
- Machine Learning at the Network Edge: A Survey
Recommendations
A top-down survey on securing IoT with machine learning: goals, recent advances and challenges
The Internet of Things (IoT) has seen it all from being just another innovation to a leading technology; it is now a binding force that interconnects various aspects of our lives. The IoT's tremendous growth is driven by emerging applications and evolving ...
Thoughts on Edge Intelligence
GLSVLSI '19: Proceedings of the 2019 on Great Lakes Symposium on VLSIMachine learning methods have exploded in the past half-dozen years. Machine learning is being applied to a huge range of problems across the spectrum of applications. Initial results relied on server-oriented computations. But many applications will ...
Machine learning-based computation offloading in edge and fog: a systematic review
AbstractToday, Mobile Cloud Computing (MCC) alone can no longer respond to the increasing volume of data and satisfy the necessary delays in real-time applications. In addition, challenges such as security, energy consumption, storage space, bandwidth, ...
Comments