Abstract
Wireless smart devices with restricted resources in the IoT may need to update DNNs due to environmental changes. In order to alleviate the computational complexity of DNN updates and stringent requirements on the size and distribution of datasets when re-training, Deep Transfer Learning (DTL) is proposed for transferring the knowledge that DNNs can learn from large standard datasets and reducing the number of DNN layers that need to be re-trained. However, previous work has rarely reconciled the needs of the computational process with the advantages of the computational platform, resulting in sub-optimal performance of the system. To address this problem, we propose a Light-weight Deployment Methodology, which targets for agile deployment of DNN re-training in resource-constraint scenarios. We design a Hybrid Precision Light-weight Strategy to distinguish the general feature extractor layer from the others so that the different light-weight mechanisms are utilized for efficient computing. Besides analyzing the system performance from aspects of information loss, computational throughput, and resource utilization, our design is able to generate a system configuration guidance with NSGA-II, which compromises DNN performance and system efficiency. The evaluation shows the throughput of the computation module guided by NSGA-II reach 39.30 and 35.04 GOPs respectively for each quantization mechanism, with the relative errors of only 2.52% and 4.60% from the theoretical values. And the model size is reduced by 34.72% without accuracy loss.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Beckmann, N., Sanchez, D.: Cache calculus: modeling caches through differential equations. IEEE Comput. Archit. Lett. 16(1), 1–5 (2015)
Cong, J., Fang, Z., Gill, M., Reinman, G.: Parade: a cycle-accurate full-system simulation platform for accelerator-rich architectural design and exploration. In: 2015 IEEE/ACM International Conference on Computer-Aided Design (ICCAD), pp. 380–387. IEEE (2015)
Drumond, M., Lin, T., Jaggi, M., Falsafi, B.: End-to-end DNN training with block floating point arithmetic. CoRR abs/1804.01526 (2018). http://arxiv.org/abs/1804.01526
Geng, T., et al.: FPDeep: acceleration and load balancing of CNN training on FPGA clusters. In: 2018 IEEE 26th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM), pp. 81–84. IEEE (2018)
Gupta, S., Agrawal, A., Gopalakrishnan, K., Narayanan, P.: Deep learning with limited numerical precision. In: International Conference on Machine Learning, pp. 1737–1746. PMLR (2015)
Han, S., Mao, H., Dally, W.J.: Deep compression: compressing deep neural networks with pruning, trained quantization and Huffman coding. arXiv preprint arXiv:1510.00149 (2015)
Micikevicius, P., et al.: Mixed precision training. CoRR abs/1710.03740 (2017). http://arxiv.org/abs/1710.03740
Sriraman, A., Dhanotia, A.: Accelerometer: understanding acceleration opportunities for data center overheads at hyperscale. In: Proceedings of the Twenty-Fifth International Conference on Architectural Support for Programming Languages and Operating Systems, pp. 733–750 (2020)
Vanhoucke, V., Senior, A., Mao, M.Z.: Improving the speed of neural networks on CPUs. In: Deep Learning and Unsupervised Feature Learning Workshop, NIPS 2011 (2011)
Yang, G., Zhang, T., Kirichenko, P., Bai, J., Wilson, A.G., De Sa, C.: SWALP: stochastic weight averaging in low precision training. In: International Conference on Machine Learning, pp. 7015–7024. PMLR (2019)
Acknowledgments
This work is supported by National Natural Science Foundation of China (No. 61772094), Chongqing Municipal Natural Science Foundation (No. cstc2020jcyj-msxmx0724), Venture & Innovation Support Program for Chongqing Overseas Returnees (cx2019094), and the Fundamental Research Funds for the Central Universities, (No. 2020cdjqy-a019, No. 2020cdj-lhzz-054, No. 2019cdxyjsj0021).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Sun, S., Guo, S., Xiao, C., Liao, Z. (2021). A Light-Weight Deployment Methodology for DNN Re-training in Resource-Constraint Scenarios. In: Liu, Z., Wu, F., Das, S.K. (eds) Wireless Algorithms, Systems, and Applications. WASA 2021. Lecture Notes in Computer Science(), vol 12937. Springer, Cham. https://doi.org/10.1007/978-3-030-85928-2_35
Download citation
DOI: https://doi.org/10.1007/978-3-030-85928-2_35
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-85927-5
Online ISBN: 978-3-030-85928-2
eBook Packages: Computer ScienceComputer Science (R0)