mlsanityData to 7 Oct 2026

Architecture Aware Latency Constrained Sparse Neural Networks

arXiv 2109.00170 ↗First posted 1 Sep 2021cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Acceleration of deep neural networks to meet a specific latency constraint is essential for their deployment on mobile devices. In this paper, we design an architecture aware latency constrained sparse (ALCS) framework to prune and accelerate CNN models. Taking modern mobile computation architectures into consideration, we propose Single Instruction Multiple Data (SIMD)-structured pruning, along with a novel sparse convolution algorithm for efficient computation. Besides, we propose to estimate the run time of sparse models with piece-wise linear interpolation. The whole latency constrained pruning task is formulated as a constrained optimization problem that can be efficiently solved with Alternating Direction Method of Multipliers (ADMM). Extensive experiments show that our system-algorithm co-design framework can achieve much better Pareto frontier among network accuracy and latency on resource-constrained mobile devices.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 46 of its 54 references that are in the map. First posted on arXiv 1 Sep 2021.

Builds on

46 of 54 references.

References · 46

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  4. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
  5. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  6. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,829
  7. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
  8. 2014Going Deeper with Convolutionscited by 4,231
  9. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,827
  10. 2015SSD: Single Shot MultiBox Detectorcited by 3,125
  11. 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,291
  12. 2013Rich feature hierarchies for accurate object detection and semantic segmentationcited by 2,160
34 more
  1. 2015Learning both Weights and Connections for Efficient Neural Networkscited by 1,998
  2. 2019Searching for MobileNetV3cited by 1,984
  3. 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,548
  4. 2021EfficientNetV2: Smaller Models and Faster Trainingcited by 1,127
  5. 2019Once-for-All: Train One Network and Specialize it for Efficient Deploymentcited by 778
  6. 2016Learning Structured Sparsity in Deep Neural Networkscited by 538
  7. 2018HAQ: Hardware-Aware Automated Quantization with Mixed Precisioncited by 426
  8. 2018AMC: AutoML for Model Compression and Acceleration on Mobile Devicescited by 375
  9. 2017Scalable Training of Artificial Neural Networks with Adaptive Sparse Connectivity inspired by Network Sciencecited by 325
  10. 2014Exploiting Linear Structure Within Convolutional Networks for Efficient Evaluationcited by 304
  11. 2020HRank: Filter Pruning using High-Rank Feature Mapcited by 291
  12. 2014Speeding up Convolutional Neural Networks with Low Rank Expansionscited by 283
  13. 2019Universally Slimmable Networks and Improved Training Techniquescited by 219
  14. 2015Quantized Convolutional Neural Networks for Mobile Devicescited by 212
  15. 2019Sparse Networks from Scratch: Faster Training without Losing Performancecited by 210
  16. 2014Speeding-up Convolutional Neural Networks Using Fine-tuned CP-Decompositioncited by 199
  17. 2019Parameter Efficient Training of Deep Convolutional Neural Networks by Dynamic Sparse Reparameterizationcited by 187
  18. 2018NetAdapt: Platform-Aware Neural Network Adaptation for Mobile Applicationscited by 146
  19. 2017Deep Rewiring: Training very sparse deep networkscited by 136
  20. 2016Designing Energy-Efficient Convolutional Neural Networks using Energy-Aware Pruningcited by 133
  21. 2020Dynamic Model Pruning with Feedbackcited by 125
  22. 2015Global Convergence of ADMM in Nonconvex Nonsmooth Optimizationcited by 98
  23. 2019Fast Sparse ConvNetscited by 86
  24. 2019Global Sparse Momentum SGD for Pruning Very Deep Neural Networkscited by 70
  25. 2020DMCP: Differentiable Markov Channel Pruning for Neural Networkscited by 68
  26. 2017Extremely Low Bit Neural Network: Squeeze the Last Bit Out with ADMMcited by 60
  27. 2020DHP: Differentiable Meta Pruning via HyperNetworkscited by 57
  28. 2020Model Rubik's Cube: Twisting Resolution, Depth and Width for TinyNetscited by 49
  29. 2020DSA: More Efficient Budgeted Pruning via Differentiable Sparsity Allocationcited by 47
  30. 2016Faster CNNs with Direct Sparse Convolutions and Guided Pruningcited by 39
  31. 2020AOWS: Adaptive and optimal network width search with latency constraintscited by 17
  32. 2018ECC: Platform-Independent Energy-Constrained Deep Neural Network Compression via a Bilinear Regression Modelcited by 15
  33. 2018Energy-Constrained Compression for Deep Neural Networks via Weighted Sparse Projection and Layer Input Maskingcited by 11
  34. 2019AutoSlim: Towards One-Shot Architecture Search for Channel Numberscited by 10

Built on by

Not cited yet.