Rapid Deployment of DNNs for Edge Computing via Structured Pruning at Initialization
Not on the map. Why?
Abstract (arXiv)
Edge machine learning (ML) enables localized processing of data on devices and is underpinned by deep neural networks (DNNs). However, DNNs cannot be easily run on devices due to their substantial computing, memory and energy requirements for delivering performance that is comparable to cloud-based ML. Therefore, model compression techniques, such as pruning, have been considered. Existing pruning methods are problematic for edge ML since they: (1) Create compressed models that have limited runtime performance benefits (using unstructured pruning) or compromise the final model accuracy (using structured pruning), and (2) Require substantial compute resources and time for identifying a suitable compressed DNN model (using neural architecture search). In this paper, we explore a new avenue, referred to as Pruning-at-Initialization (PaI), using structured pruning to mitigate the above problems. We develop Reconvene, a system for rapidly generating pruned models suited for edge deployments using structured PaI. Reconvene systematically identifies and prunes DNN convolution layers that are least sensitive to structured pruning. Reconvene rapidly creates pruned DNNs within seconds that are up to 16.21x smaller and 2x faster while maintaining the same accuracy as an unstructured PaI counterpart.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 34 of its 47 references that are in the map. First posted on arXiv 22 Apr 2024.
Builds on
34 of 47 references.
References · 34
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
- 2015Distilling the Knowledge in a Neural Networkcited by 10,076
- 2018The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networkscited by 2,250
- 2015Learning both Weights and Connections for Efficient Neural Networkscited by 1,998
- 2018DARTS: Differentiable Architecture Searchcited by 1,782
- 2016Pruning Filters for Efficient ConvNetscited by 1,129
- 2017Learning Transferable Architectures for Scalable Image Recognitioncited by 1,097
- 2019Once-for-All: Train One Network and Specialize it for Efficient Deploymentcited by 778
- 2017Channel Pruning for Accelerating Very Deep Neural Networkscited by 697
- 2018Rethinking the Value of Network Pruningcited by 595
- 2020What is the State of Neural Network Pruning?cited by 578
22 more
- 2016Learning Structured Sparsity in Deep Neural Networkscited by 538
- 2021RepVGG: Making VGG-style ConvNets Great Againcited by 520
- 2020Pruning neural networks without any data by iteratively conserving synaptic flowcited by 475
- 2019The State of Sparsity in Deep Neural Networkscited by 471
- 2020Picking Winning Tickets Before Training by Preserving Gradient Flowcited by 467
- 2016EIE: Efficient Inference Engine on Compressed Deep Neural Networkcited by 291
- 2020Neural Architecture Search without Trainingcited by 241
- 2019Model Pruning Enables Efficient Federated Learning on Edge Devicescited by 219
- 2023DepGraph: Towards Any Structural Pruningcited by 202
- 2020Pruning Neural Networks at Initialization: Why are We Missing the Mark?cited by 181
- 2019Convergence of Edge Computing and Deep Learning: A Comprehensive Surveycited by 120
- 2022ZeroFL: Efficient On-Device Training for Federated Learning with Local Sparsitycited by 45
- 2022Prospect Pruning: Finding Trainable Weights at Initialization using Meta-Gradientscited by 34
- 2019Non-Structured DNN Weight Pruning -- Is It Beneficial in Any Platform?cited by 31
- 2021Personalized Federated Learning by Structured and Unstructured Pruning under Data Heterogeneitycited by 27
- 2016Challenges and Opportunities in Edge Computingcited by 26
- 2020Single Shot Structured Pruning Before Trainingcited by 19
- 2023Why is the State of Neural Network Pruning so Confusing? On the Fairness, Comparison Setup, and Trainability in Network Pruningcited by 16
- 2018Characterising Across-Stack Optimisations for Deep Convolutional Neural Networkscited by 10
- 2023DNNShifter: An Efficient DNN Pruning System for Edge Computingcited by 8
- 2022Structured Pruning is All You Need for Pruning CNNs at Initializationcited by 7
- 2021Heterogeneous Federated Learning using Dynamic Model Pruning and Adaptive Gradientcited by 4
Built on by
Citing papers per year: 2025: 1.
1 paper.