Shreshth Tuli
Abstract
In recent years, deep learning models have become ubiquitous in industry and academia alike. Modern deep neural networks can solve one of the most complex problems today, but coming with the price of massive compute and storage requirements. This makes deploying such massive neural networks challenging in the mobile edge computing paradigm, where edge nodes are resource-constrained, hence limiting the input analysis power of such frameworks. Semantic and layer-wise splitting of neural networks for distributed processing show some hope in this direction. However, there are no intelligent algorithms that place such modular splits to edge nodes for optimal performance. This work proposes a novel placement policy, SplitPlace, for the placement of such neural network split fragments on mobile edge hosts for efficient and scalable computing.
- Nasir Abbas et al. "Mobile edge computing: A survey". In: IEEE Internet of Things Journal 5.1 (2017), pp. 450--465.Google Scholar
Cross Ref
- Ejaz Ahmed and Mubashir Husain Rehmani. "Mobile Edge Computing: Opportunities, solutions, and challenges". In: Future Generation Computer Systems 70 (2017), pp. 59--63. issn: 0167--739X.Google ScholarDigital Library
- Md Zahangir Alom et al. "The history began from alexnet: A comprehensive survey on deep learning approaches". In: arXiv preprint arXiv:1803.01164 (2018).Google Scholar
- Alessandro Capotondi et al. "CMix-NN: Mixed lowprecision CNN library for memory-constrained edge devices". In: IEEE Transactions on Circuits and Systems II: Express Briefs 67.5 (2020), pp. 871--875.Google ScholarCross Ref
- Jiasi Chen and Xukan Ran. "Deep LearningWith Edge Computing: A Review." In: Proceedings of the IEEE 107.8 (2019), pp. 1655--1674.Google ScholarCross Ref
- Amir Erfan Eshratifar, Amirhossein Esmaili, and Massoud Pedram. "Bottlenet: A deep learning architecture for intelligent mobile cloud computing services". In: 2019 IEEE/ACM International Symposium on Low Power Electronics and Design (ISLPED). IEEE. 2019, pp. 1--6.Google ScholarCross Ref
- Sukhpal Singh Gill et al. "Transformative effects of IoT, Blockchain and Artificial Intelligence on cloud computing: Evolution, vision, trends and open challenges". In: Internet of Things (2019), pp. 100--118.Google Scholar
- V Scott Gordon and Jeb Crouson. "Self-splitting modular neural network-domain partitioning at boundaries of trained regions". In: International Joint Conference on Neural Networks. IEEE. 2008, pp. 1085--1091.Google Scholar
- Jashwant Raj Gunasekaran et al. "Implications of Public Cloud Resource Heterogeneity for Inference Serving". In: Proceedings of the 2020 Sixth International Workshop on Serverless Computing. 2020, pp. 7--12.Google Scholar
- Juyong Kim et al. "SplitNet: Learning to semantically split deep networks for parameter reduction and model parallelization". In: International Conference on Machine Learning (ICML). JMLR. 2017, pp. 1866--1874.Google Scholar
- Quan Le et al. "An overview of deep learning in industry". In: Data Analytics and AI (2020), pp. 65-- 98.Google Scholar
- Wei Yang Bryan Lim et al. "Federated learning in mobile edge networks: A comprehensive survey". In: IEEE Communications Surveys & Tutorials 22.3 (2020), pp. 2031--2063.Google ScholarCross Ref
- Yuyi Mao, Jun Zhang, and Khaled B Letaief. "Dynamic computation offloading for mobile-edge computing with energy harvesting devices". In: IEEE Journal on Selected Areas in Communications 34.12 (2016), pp. 3590--3605.Google ScholarDigital Library
- Yoshitomo Matsubara et al. "Distilled Split Deep Neural Networks for Edge-Assisted Real-Time Systems". In: Workshop on Hot Topics in Video Analytics and Intelligent Edges. 2019, pp. 21--26.Google Scholar
- Yuanming Shi et al. "Communication-efficient edge AI: Algorithms and systems". In: IEEE Communications Surveys & Tutorials 22.4 (2020), pp. 2167--2191.Google ScholarCross Ref
- Surat Teerapittayanon, BradleyMcDanel, and Hsiang- Tsung Kung. "Distributed deep neural networks over the cloud, the edge and end devices". In: 2017 IEEE 37th International Conference on Distributed Computing Systems (ICDCS). IEEE. 2017, pp. 328--339.Google ScholarCross Ref
- Shreshth Tuli et al. "COSCO: Container Orchestration using Co-Simulation and Gradient Based Optimization for Fog Computing Environments". In: IEEE Transactions on Parallel and Distributed Systems (2021).Google Scholar
- Shreshth Tuli et al. "Dynamic Scheduling for Stochastic Edge-Cloud Computing Environments using A3C learning and Residual Recurrent Neural Networks". In: IEEE Transactions on Mobile Computing (2020).Google Scholar
- Shreshth Tuli et al. "Fogbus: A blockchain-based lightweight framework for edge and fog computing". In: Journal of Systems and Software (2019).Google Scholar
- Shreshth Tuli et al. "HealthFog: An ensemble deep learning based smart healthcare system for automatic diagnosis of heart diseases in integrated IoT and fog computing environments". In: Future Generation Computer Systems 104 (2020), pp. 187--200.Google ScholarCross Ref
- Shreshth Tuli et al. "iThermoFog: IoT-Fog based automatic thermal profile creation for cloud data centers using artificial intelligence techniques". In: Internet Technology Letters 3.5 (2020), e198.Google Scholar
- Yury A Ushakov et al. "Split Neural Networks for Mobile Devices". In: Telecommunications Forum (TELFOR). IEEE. 2018, pp. 420--425.Google Scholar
- Minchen Yu et al. "Gillis: Serving Large Neural Networks in Serverless Functions with Automatic Model Partitioning". In: 41st IEEE International Conference on Distributed Computing Systems. 2021.Google Scholar
- Hongyu Zhu et al. "TBD: Benchmarking and analyzing deep neural network training". In: arXiv preprint arXiv:1803.06905 (2018).Google Scholar
Index Terms
- SplitPlace: Intelligent Placement of Split Neural Nets in Mobile Edge Environments
Recommendations
SplitPlace: AI Augmented Splitting and Placement of Large-Scale Neural Networks in Mobile Edge Environments
In recent years, deep learning models have become ubiquitous in industry and academia alike. Deep neural networks can solve some of the most complex pattern-recognition problems today, but come with the price of massive compute and memory requirements. ...
Multiplicative neuron model artificial neural network based on Gaussian activation function
Multiplicative neuron model-based artificial neural networks are one of the artificial neural network types which have been proposed recently and have produced successful forecasting results. Sigmoid activation function was used in multiplicative neuron ...
Identification and control using MLP, Elman, NARXSP and radial basis function networks: a comparative analysis
This paper describes four neural networks multilayer perceptron (MLP) network, Elman network, NARXSP network and radial basis function (RBF) network. Neural networks are applied for identification and control of DC servo motor and benchmark nonlinear ...
Comments