NestDNN: Resource-Aware Multi-Tenant On-Device Deep Learning for Continuous Mobile Vision
Open accessAbstract
Mobile vision systems such as smartphones, drones, and augmented-reality headsets are revolutionizing our lives. These systems usually run multiple applications concurrently and their available resources at runtime are dynamic due to events such as starting new applications, closing existing applications, and application priority changes. In this paper, we present NestDNN, a framework that takes the dynamics of runtime resources into account to enable resource-aware multi-tenant on-device deep learning for mobile vision systems. NestDNN enables each deep learning model to offer flexible resource-accuracy trade-offs. At runtime, it dynamically selects the optimal resource-accuracy trade-off for each deep learning model to fit the model's resource demand to the system's available runtime resources. In doing so, NestDNN efficiently utilizes the limited resources in mobile vision systems to jointly maximize the performance of all the concurrently running applications. Our experiments show that compared to the resource-agnostic status quo approach, NestDNN achieves as much as 4.2% increase in inference accuracy, 2.0× increase in video frame processing rate and 1.7× reduction on energy consumption.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
NestDNN transforms the DL model into a set of descendant models, in which the descendant model with fewer resource requirements shares its model parameters with the descendant model requiring more resources, making itself nested inside the descendent model requiring more resources without taking extra memory space.
“With the model pruning and recovery scheme, NestDNN [105] transforms the DL model into a set of descendant models, in which the descendant model with fewer resource requirements shares its model parameters with the descendant model requiring more resources, making itself nested inside the descendent model requiring more resources without taking extra memory space.”
From Convergence of Edge Computing and Deep Learning: A Comprehensive Survey (2020) · cites this paper as [105]
NestDNN's multi-capacity model provides variable resource-accuracy trade-offs with a compact memory footprint, ensuring efficient multi-tenant DL at the end level.
“In this way, the multi-capacity model provides variable resourceaccuracy trade-offs with a compact memory footprint, hence ensuring efficient multi-tenant DL at the end level.”
From Convergence of Edge Computing and Deep Learning: A Comprehensive Survey (2020) · cites this paper as [105]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
The citation lists haven’t been fetched yet
References 16 · Cited by 291
Research works printed but not linked: 15
- [6]6. Paramvir Bahl, Matthai Philipose, and Lin Zhong. 2012. Vision: cloud- powered sight for all: showing the cloud what you see. In ACM work- shop on Mobile cloud computing and services. 53-60.
- [8]8. Biyi Fang, Jillian Co, and Mi Zhang. 2017. DeepASL: Enabling Ubiq- uitous and Non-Intrusive Word and Sentence-Level Sign Language Translation. In ACM SenSys. Delft, The Netherlands.
- [9]9. Yiwen Guo, Anbang Yao, and Yurong Chen. 2016. Dynamic network surgery for efficient dnns. In NIPS. 1379-1387.
- [11]11. Song Han, Jeff Pool, John Tran, and William Dally. 2015. Learning both weights and connections for efficient neural network. In NIPS. 1135-1143.
- [12]12. Seungyeop Han, Haichen Shen, Matthai Philipose, Sharad Agar- wal, Alec Wolman, and Arvind Krishnamurthy. 2016. MCDNN: An Approximation-Based Execution Framework for Deep Stream Process- ing Under Resource Constraints. In ACM MobiSys. 123-136.
- [15]15. Gao Huang, Zhuang Liu, Kilian Q Weinberger, and Laurens van der Maaten. 2017. Densely connected convolutional networks. In IEEE conference on computer vision and pattern recognition, Vol. 1. 3.
- [19]19. Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton. 2012. Imagenet classification with deep convolutional neural networks. In NIPS. 1097- 1105.
- [20]20. Nicholas D Lane, Sourav Bhattacharya, Petko Georgiev, Claudio For- livesi, Lei Jiao, Lorena Qendro, and Fahim Kawsar. 2016. DeepX: A software accelerator for low-power deep learning inference on mobile devices. In ACM/IEEE IPSN. 1-12.
- [21]21. Yann LeCun, Yoshua Bengio, and Geoffrey Hinton. 2015. Deep learning. Nature 521, 7553 (2015), 436-444.
- [24]24. Robert LiKamWa and Lin Zhong. 2015. Starfish: Efficient concurrency support for computer vision applications. In ACM MobiSys. 213-226.
- [25]25. Ji Lin, Yongming Rao, Jiwen Lu, and Jie Zhou. 2017. Runtime neural pruning. In NIPS. 2181-2191.
- [29]29. Saman Naderiparizi, Pengyu Zhang, Matthai Philipose, Bodhi Priyan- tha, Jie Liu, and Deepak Ganesan. 2017. Glimpse: A Programmable Early-Discard Camera Architecture for Continuous Mobile Vision. In ACM MobiSys. 292-305.
- [39]39. Haoyu Zhang, Ganesh Ananthanarayanan, Peter Bodik, Matthai Phili- pose, Paramvir Bahl, and Michael J Freedman. 2017. Live Video Ana- lytics at Scale with Approximation and Delay-Tolerance.. In NSDI.
- [40]40. Bolei Zhou, Agata Lapedriza, Aditya Khosla, Aude Oliva, and Antonio Torralba. 2017. Places: A 10 million Image Database for Scene Recog- nition. IEEE Transactions on Pattern Analysis and Machine Intelligence (2017).
- [41]41. Bolei Zhou, Agata Lapedriza, Jianxiong Xiao, Antonio Torralba, and Aude Oliva. 2014. Learning deep features for scene recognition using places database. In NIPS. 487-495.
- Authors
- Fang, B., Zeng, X., & Zhang, M.
- Affiliations
- Michigan State University (US)
- Venue
- Proceedings of the 24th Annual International Conference on Mobile Computing and Networking
- Publisher
- ACM
- Published
- 2018-10-15
- Pages
- 115-127
- ISSN
- 1543-5679 (ISSN-L)
- OpenAlex
- W2897268228
- Cited by
- 291
- References
- 16
- Citations by year
- 2026: 13 · 2025: 29 · 2024: 41 · 2023: 36 · 2022: 46 · 2021: 53 · 2020: 50 · 2019: 20 · 2018: 3
- FWCI
- 9.49
- Type
- conference paper
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- ACM Annual International Conference on Mobile Computing and Networking (MobiCom) (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Science Foundation (1617627, 1565604, 1632051)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Hardware and Architecture › Parallel Computing and Optimization Techniques
- Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing
- Keywords
- on-device deep learning, mobile vision systems, energy efficiency, inference accuracy
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
