Please use this identifier to cite or link to this item:
|Title:||ZeroBN : learning compact neural networks for latency-critical edge systems||Authors:||Huai, Shuo
|Keywords:||Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence||Issue Date:||2021||Source:||Huai, S., Zhang, L., Liu, D., Liu, W. & Subramaniam, R. (2021). ZeroBN : learning compact neural networks for latency-critical edge systems. 2021 58th ACM/IEEE Design Automation Conference (DAC), 151-156. https://dx.doi.org/10.1109/DAC18074.2021.9586309||Project:||I1801E0028||Abstract:||Edge devices have been widely adopted to bring deep learning applications onto low power embedded systems, mitigating the privacy and latency issues of accessing cloud servers. The increasingly computational demand of complex neural network models leads to large latency on edge devices with limited resources. Many application scenarios are real-time and have a strict latency constraint, while conventional neural network compression methods are not latency-oriented. In this work, we propose a novel compact neural networks training method to reduce the model latency on latency-critical edge systems. A latency predictor is also introduced to guide and optimize this procedure. Coupled with the latency predictor, our method can guarantee the latency for a compact model by only one training process. The experiment results show that, compared to state-of-the-art model compression methods, our approach can well-fit the 'hard' latency constraint by significantly reducing the latency with a mild accuracy drop. To satisfy a 34ms latency constraint, we compact ResNet-50 with 0.82% of accuracy drop. And for GoogLeNet, we can even increase the accuracy by 0.3%||URI:||https://hdl.handle.net/10356/155572||ISBN:||9781665432740||DOI:||10.1109/DAC18074.2021.9586309||DOI (Related Dataset):||10.21979/N9/IRNJ4I||Rights:||©2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The published version is available at: https://doi.org/10.1109/DAC18074.2021.9586309.||Fulltext Permission:||open||Fulltext Availability:||With Fulltext|
|Appears in Collections:||SCSE Conference Papers|
Updated on Jul 9, 2022
Updated on Sep 28, 2022
Updated on Sep 28, 2022
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.