Papers
Topics
Authors
Recent
Assistant
AI Research Assistant
Well-researched responses based on relevant abstracts and paper content.
Custom Instructions Pro
Preferences or requirements that you'd like Emergent Mind to consider when generating responses.
GPT-5.1
GPT-5.1 114 tok/s
Gemini 3.0 Pro 53 tok/s Pro
Gemini 2.5 Flash 132 tok/s Pro
Kimi K2 176 tok/s Pro
Claude Sonnet 4.5 37 tok/s Pro
2000 character limit reached

SC2 Benchmark: Supervised Compression for Split Computing (2203.08875v2)

Published 16 Mar 2022 in cs.LG, cs.CV, and eess.IV

Abstract: With the increasing demand for deep learning models on mobile devices, splitting neural network computation between the device and a more powerful edge server has become an attractive solution. However, existing split computing approaches often underperform compared to a naive baseline of remote computation on compressed data. Recent studies propose learning compressed representations that contain more relevant information for supervised downstream tasks, showing improved tradeoffs between compressed data size and supervised performance. However, existing evaluation metrics only provide an incomplete picture of split computing. This study introduces supervised compression for split computing (SC2) and proposes new evaluation criteria: minimizing computation on the mobile device, minimizing transmitted data size, and maximizing model accuracy. We conduct a comprehensive benchmark study using 10 baseline methods, three computer vision tasks, and over 180 trained models, and discuss various aspects of SC2. We also release sc2bench, a Python package for future research on SC2. Our proposed metrics and package will help researchers better understand the tradeoffs of supervised compression in split computing.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (94)
  1. Understanding the Limits of LoRaWAN. IEEE Communications Magazine, 55(9):34–40, 2017.
  2. Scale-space flow for end-to-end optimized video compression. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  8503–8512, 2020.
  3. Neural Rate Estimator and Unsupervised Learning for Efficient Distributed Image Analytics in Split-DNN Models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  2022–2030, 2023.
  4. Deep Variational Information Bottleneck. In International Conference on Learning Representations, 2017.
  5. Pareto-Optimal Bit Allocation for Collaborative Intelligence. IEEE Transactions on Image Processing, 30:3348–3361, 2021.
  6. Collaborative Object Detectors Adaptive to Bandwidth and Computation. In ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp.  2839–2843. IEEE, 2022.
  7. Density Modeling of Images using a Generalized Normalization Transformation. In International Conference on Learning Representations, 2016.
  8. End-to-end Optimized Image Compression. International Conference on Learning Representations, 2017.
  9. Variational image compression with a scale hyperprior. In International Conference on Learning Representations, 2018.
  10. Auto-Split: A General Framework of Collaborative Edge-Cloud AI. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, pp.  2543–2553, 2021.
  11. CompressAI: a PyTorch library and evaluation platform for end-to-end compression research. arXiv preprint arXiv:2011.03029, 2020. https://github.com/InterDigitalInc/CompressAI.
  12. Fabrice Bellard. BPG Image format. https://bellard.org/bpg/ [Accessed on February 18, 2023].
  13. Deep Learning With Edge Computing: A Review. Proceedings of the IEEE, 107(8):1655–1674, 2019.
  14. DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs. IEEE Transactions on Pattern Analysis and Machine Intelligence, 40(4):834–848, 2017a.
  15. Rethinking Atrous Convolution for Semantic Image Segmentation. arXiv preprint arXiv:1706.05587, 2017b.
  16. Learning for Video Compression. IEEE Transactions on Circuits and Systems for Video Technology, 30(2):566–576, 2019.
  17. Learned Image Compression With Discretized Gaussian Mixture Likelihoods and Attention Modules. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  7939–7948, 2020.
  18. Back-And-Forth Prediction for Deep Tensor Compression. In ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp.  4467–4471. IEEE, 2020.
  19. Task-aware quantization network for jpeg image compression. In European Conference on Computer Vision, pp.  309–324. Springer, 2020.
  20. A Low-Complexity Approach to Rate-Distortion Optimized Variable Bit-Rate Compression for Split DNN Computing. In 2022 26th International Conference on Pattern Recognition (ICPR), pp.  182–188. IEEE, 2022.
  21. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pp.  4171–4186, 2019.
  22. Neural inter-frame compression for video coding. In 2019 IEEE/CVF International Conference on Computer Vision (ICCV), pp.  6420–6428, 2019.
  23. SplitNets: Designing Neural Architectures for Efficient Distributed Computing on Head-Mounted Systems. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  12559–12569, 2022.
  24. An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale. In International Conference on Learning Representations, 2021.
  25. Lossy Compression for Lossless Prediction. In Neural Compression: From Information Theory to Applications–Workshop@ ICLR 2021, 2021.
  26. JointDNN: an efficient training and inference engine for intelligent mobile cloud computing services. IEEE Transactions on Mobile Computing, 2019a.
  27. BottleNet: A Deep Learning Architecture for Intelligent Mobile Cloud Computing Services. In 2019 IEEE/ACM Int. Symposium on Low Power Electronics and Design (ISLPED), pp.  1–6, 2019b.
  28. The PASCAL Visual Object Classes Challenge 2012 (VOC2012). 2012.
  29. Compression without Quantization. In OpenReview, 2019.
  30. Google. Compression Techniques | WebP | Google Developers. https://developers.google.com/speed/webp/docs/compression [Accessed on February 18, 2023].
  31. Distributed learning of deep neural network over multiple agents. Journal of Network and Computer Applications, 116:1–8, 2018.
  32. Activation Sparsity and Dynamic Pruning for Split Computing in Edge AI. In Proceedings of the 3rd International Workshop on Distributed Machine Learning, pp.  30–36, 2022.
  33. Video compression with rate-distortion autoencoders. In Proceedings of the IEEE International Conference on Computer Vision, pp.  7033–7042, 2019.
  34. Deep generative video compression. In Advances in Neural Information Processing Systems, volume 32, 2019.
  35. ELIC: Efficient Learned Image Compression With Unevenly Grouped Space-Channel Contextual Adaptive Coding. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  5718–5727, 2022.
  36. Deep Residual Learning for Image Recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.  770–778, 2016.
  37. Mask R-CNN. In Proceedings of the IEEE International Conference on Computer Vision, pp.  2961–2969, 2017.
  38. Rethinking ImageNet Pre-training. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pp.  4918–4927, 2019.
  39. Distilling the Knowledge in a Neural Network. In Deep Learning and Representation Learning Workshop: NIPS 2014, 2014.
  40. Searching for MobileNetV3. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pp.  1314–1324, 2019.
  41. Densely Connected Convolutional Networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.  4700–4708, 2017.
  42. Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.  2704–2713, 2018.
  43. Improved Lossy Image Compression with Priming and Spatially Adaptive Bit Rates for Recurrent Networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.  4385–4393, 2018.
  44. Neurosurgeon: Collaborative Intelligence Between the Cloud and Mobile Edge. In Proceedings of the Twenty-Second International Conference on Architectural Support for Programming Languages and Operating Systems, pp. 615–629, 2017. ISBN 978-1-4503-4465-4. doi: 10.1145/3037697.3037698.
  45. Adam: A Method for Stochastic Optimization. In Third International Conference on Learning Representations, 2015.
  46. Auto-Encoding Variational Bayes. In International Conference on Learning Representations, 2014.
  47. Alex Krizhevsky. Learning Multiple Layers of Features from Tiny Images. 2009.
  48. Gradient-Based Learning Applied to Document Recognition. Proceedings of the IEEE, 86(11):2278–2324, 1998.
  49. Microsoft COCO: Common Objects in Context. In European conference on computer vision, pp.  740–755. Springer, 2014.
  50. Feature Pyramid Networks for Object Detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.  2117–2125, 2017a.
  51. Focal loss for dense object detection. In Proceedings of the IEEE International Conference on Computer Vision, pp.  2980–2988, 2017b.
  52. Lossless Compression with Probabilistic Circuits. In International Conference on Learning Representations, 2022.
  53. SGDR: Stochastic Gradient Descent with Warm Restarts. In International Conference on Learning Representations, 2017.
  54. Dvc: An end-to-end deep video compression framework. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.  11006–11015, 2019.
  55. Progressive Neural Image Compression with Nested Quantization and Latent Ordering. In 2021 IEEE International Conference on Image Processing (ICIP), pp.  539–543. IEEE, 2021.
  56. Yoshitomo Matsubara. torchdistill: A Modular, Configuration-Driven Framework for Knowledge Distillation. In International Workshop on Reproducible Research in Pattern Recognition, pp.  24–44. Springer, 2021. https://github.com/yoshitomo-matsubara/torchdistill.
  57. Neural Compression and Filtering for Edge-assisted Real-time Object Detection in Challenged Networks. In 2020 25th International Conference on Pattern Recognition (ICPR), pp.  2272–2279, 2021.
  58. Distilled Split Deep Neural Networks for Edge-Assisted Real-Time Systems. In Proceedings of the 2019 Workshop on Hot Topics in Video Analytics and Intelligent Edges, pp.  21–26, 2019.
  59. Head Network Distillation: Splitting Distilled Deep Neural Networks for Resource-Constrained Edge Computing Systems. IEEE Access, 8:212177–212193, 2020. doi: 10.1109/ACCESS.2020.3039714.
  60. BottleFit: Learning Compressed Representations in Deep Neural Networks for Effective and Efficient Split Computing. In 2022 IEEE 23rd International Symposium on a World of Wireless, Mobile and Multimedia Networks (WoWMoM), pp.  337–346. IEEE, 2022a.
  61. Split Computing and Early Exiting for Deep Learning Applications: Survey and Research Challenges. ACM Computing Surveys, 55(5):1–30, 2022b.
  62. Supervised Compression for Resource-Constrained Edge Computing Systems. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp.  2685–2695, 2022c.
  63. Channel-Wise Autoregressive Entropy Models for Learned Image Compression. In 2020 IEEE International Conference on Image Processing (ICIP), pp.  3339–3343. IEEE, 2020.
  64. Joint Autoregressive and Hierarchical Priors for Learned Image Compression. In Advances in Neural Information Processing Systems, pp. 10771–10780, 2018.
  65. Picture coding: A review. Proceedings of the IEEE, 68(3):366–406, 1980.
  66. PyTorch: An imperative style, high-performance deep learning library. In Advances in Neural Information Processing Systems, pp. 8024–8035, 2019.
  67. Learning Transferable Visual Models From Natural Language Supervision. arXiv preprint arXiv:2103.00020, 2021.
  68. Designing Network Design Spaces. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  10428–10436, 2020.
  69. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. In Advances in Neural Information Processing Systems, pp. 91–99, 2015.
  70. ImageNet Large Scale Visual Recognition Challenge. International Journal of Computer Vision, 115(3):211–252, 2015.
  71. IoT Technologies for Embedded Computing: A Survey. In 2016 International Conference on Hardware/Software Codesign and System Synthesis (CODES+ ISSS), pp.  1–10. IEEE, 2016.
  72. MobileNetV2: Inverted Residuals and Linear Bottlenecks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.  4510–4520, 2018.
  73. BottleNet++: An end-to-end approach for feature compression in device-edge co-inference systems. In 2020 IEEE International Conference on Communications Workshops (ICC Workshops), pp.  1–6. IEEE, 2020.
  74. End-to-end Learning of Compressible Features. In 2020 IEEE International Conference on Image Processing (ICIP), pp.  3349–3353. IEEE, 2020.
  75. How to train your ViT? Data, Augmentation, and Regularization in Vision Transformers. Transactions on Machine Learning Research, 2022. ISSN 2835-8856. URL https://openreview.net/forum?id=4nPswr1KcP.
  76. MnasNet: Platform-Aware Neural Architecture Search for Mobile. In Proceedings of the IEEE Conf. on Computer Vision and Pattern Recognition, pp.  2820–2828, 2019.
  77. Lossy Image Compression with Compressive Autoencoders. In International Conference on Learning Representations, 2017.
  78. Full Resolution Image Compression with Recurrent Neural Networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.  5306–5314, 2017.
  79. Fixing the train-test resolution discrepancy. Advances in Neural Information Processing Systems, 32:8252–8262, 2019.
  80. Split learning for health: Distributed deep learning without sharing raw patient data. arXiv preprint arXiv:1812.00564, 2018.
  81. Neural Data-Dependent Transform for Learned Image Compression. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  17379–17388, 2022.
  82. Ross Wightman. Pytorch image models. https://github.com/rwightman/pytorch-image-models, 2019.
  83. Paul A Wintz. Transform picture coding. Proceedings of the IEEE, 60(7):809–820, 1972.
  84. Video compression through image interpolation. In Proceedings of the European Conference on Computer Vision (ECCV), pp.  416–431, 2018.
  85. Lossy Image Compression with Conditional Diffusion Models. arXiv preprint arXiv:2209.06950, 2022a.
  86. Insights from generative modeling for neural video compression. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023a.
  87. Y Yang and S Mandt. Towards empirical sandwich bounds on the rate-distortion function. In International Conference on Learning Representations, 2022b.
  88. Asymmetrically-powered neural image compression with shallow decoders. arXiv preprint arXiv:2304.06244, 2023.
  89. Improving Inference for Neural Image Compression. In Advances in Neural Information Processing Systems, volume 33, pp.  573–584, 2020a.
  90. Variational Bayesian Quantization. In International Conference on Machine Learning, pp. 10670–10680. PMLR, 2020b.
  91. An Introduction to Neural Data Compression. Foundations and Trends® in Computer Graphics and Vision, 15(2):113–200, 2023b.
  92. Feature Compression for Rate Constrained Object Detection on the Edge. In 2022 IEEE 5th International Conference on Multimedia Information Processing and Retrieval (MIPR), pp.  1–6. IEEE, 2022.
  93. New Flexible Multiple-Precision Multiply-Accumulate Unit for Deep Neural Network Training and Inference. IEEE Transactions on Computers, 69(1):26–38, 2019.
  94. Transformer-based Transform Coding. In International Conference on Learning Representations, 2021.
Citations (12)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
Dice Question Streamline Icon: https://streamlinehq.com

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now
Lightbulb Streamline Icon: https://streamlinehq.com

Continue Learning

We haven't generated follow-up questions for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now
List To Do Tasks Checklist Streamline Icon: https://streamlinehq.com

Collections

Sign up for free to add this paper to one or more collections.

User Circle Single Streamline Icon: https://streamlinehq.com Sign Up