Skip to main content

Meta-GF: Training Dynamic-Depth Neural Networks Harmoniously

  • Conference paper
  • First Online:
Computer Vision – ECCV 2022 (ECCV 2022)

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 13671))

Included in the following conference series:

Abstract

Most state-of-the-art deep neural networks use static inference graphs, which makes it impossible for such networks to dynamically adjust the depth or width of the network according to the complexity of the input data. Different from these static models, depth-adaptive neural networks, e.g. the multi-exit networks, aim at improving the computation efficiency by conducting adaptive inference conditioned on the input. To achieve adaptive inference, multiple output exits are attached at different depths of the multi-exit networks. Unfortunately, these exits usually interfere with each other in the training stage. The interference would reduce performance of the models and cause negative influences on the convergence speed. To address this problem, we investigate the gradient conflict of these multi-exit networks, and propose a novel meta-learning based training paradigm namely Meta-GF (meta gradient fusion) to harmoniously train these exits. Different from existing approaches, Meta-GF takes account of the importances of the shared parameters to each exit, and fuses the gradients of each exit by the meta-learned weights. Experimental results on CIFAR and ImageNet verify the effectiveness of the proposed method. Furthermore, the proposed Meta-GF requires no modification on the network structures and can be directly combined with previous training techniques. The code is available at https://github.com/SYVAE/MetaGF.

Y. Sun and J. Li—Co-first authors with equal contributions.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 89.00
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 119.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

References

  1. Bakhtiarnia, A., Zhang, Q., Iosifidis, A.: Multi-exit vision transformer for dynamic inference (2021)

    Google Scholar 

  2. Bolukbasi, T., Wang, J., Dekel, O., Saligrama, V.: Adaptive neural networks for efficient inference. In: International Conference on Machine Learning, pp. 527–536. PMLR (2017)

    Google Scholar 

  3. Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., Zagoruyko, S.: End-to-end object detection with transformers. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12346, pp. 213–229. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58452-8_13

    Chapter  Google Scholar 

  4. Chen, X., Dai, H., Li, Y., Gao, X., Song, L.: Learning to stop while learning to predict. In: International Conference on Machine Learning, pp. 1520–1530. PMLR (2020)

    Google Scholar 

  5. Chen, Y., Dai, X., Liu, M., Chen, D., Yuan, L., Liu, Z.: Dynamic convolution: attention over convolution kernels. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 11030–11039 (2020)

    Google Scholar 

  6. Chen, Z., Badrinarayanan, V., Lee, C.Y., Rabinovich, A.: Gradnorm: gradient normalization for adaptive loss balancing in deep multitask networks. In: International Conference on Machine Learning, pp. 794–803. PMLR (2018)

    Google Scholar 

  7. Dai, X., Kong, X., Guo, T.: Epnet: learning to exit with flexible multi-branch network. In: Proceedings of the 29th ACM International Conference on Information & Knowledge Management, pp. 235–244 (2020)

    Google Scholar 

  8. Dosovitskiy, A., et al.: An image is worth 16x16 words: transformers for image recognition at scale. In: International Conference on Learning Representations (2020)

    Google Scholar 

  9. Du, S., Lee, J.: On the power of over-parametrization in neural networks with quadratic activation. In: International Conference on Machine Learning, pp. 1329–1338. PMLR (2018)

    Google Scholar 

  10. Duggal, R., Freitas, S., Dhamnani, S., Chau, D.H., Sun, J.: Elf: an early-exiting framework for long-tailed classification. arXiv preprint arXiv:2006.11979 (2020)

  11. Elbayad, M., Gu, J., Grave, E., Auli, M.: Depth-adaptive transformer. In: ICLR 2020-Eighth International Conference on Learning Representations, pp. 1–14 (2020)

    Google Scholar 

  12. Fan, A., Grave, E., Joulin, A.: Reducing transformer depth on demand with structured dropout. In: International Conference on Learning Representations (2019)

    Google Scholar 

  13. Farhadi, A., Redmon, J.: Yolov3: an incremental improvement. In: Computer Vision and Pattern Recognition, pp. 1804–2767. Springer, Heidelberg (2018)

    Google Scholar 

  14. Gao, X., Zhao, Y., Dudziak, Ł., Mullins, R., Xu, C.z.: Dynamic channel pruning: feature boosting and suppression. In: International Conference on Learning Representations (2018)

    Google Scholar 

  15. Guo, M., Haque, A., Huang, D.A., Yeung, S., Fei-Fei, L.: Dynamic task prioritization for multitask learning. In: Proceedings of the European Conference on Computer Vision (ECCV), pp. 270–287 (2018)

    Google Scholar 

  16. Guo, S., Alvarez, J.M., Salzmann, M.: Expandnets: linear over-parameterization to train compact convolutional networks. In: Advances in Neural Information Processing Systems, vol. 33 (2020)

    Google Scholar 

  17. Han, Y., Huang, G., Song, S., Yang, L., Wang, H., Wang, Y.: Dynamic neural networks: a survey. IEEE Trans. Pattern Anal. Mach. Intell. (2021). https://doi.org/10.1109/TPAMI.2021.3117837

  18. He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016)

    Google Scholar 

  19. Huang, C., Lucey, S., Ramanan, D.: Learning policies for adaptive tracking with deep feature cascades. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 105–114 (2017)

    Google Scholar 

  20. Huang, G., Chen, D., Li, T., Wu, F., van der Maaten, L., Weinberger, K.Q.: Multi-scale dense networks for resource efficient image classification. arXiv preprint arXiv:1703.09844 (2017)

  21. Huang, G., Sun, Yu., Liu, Z., Sedra, D., Weinberger, K.Q.: Deep networks with stochastic depth. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9908, pp. 646–661. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46493-0_39

    Chapter  Google Scholar 

  22. Javaloy, A., Valera, I.: Rotograd: gradient homogenization in multi-task learning (2021)

    Google Scholar 

  23. Jeon, G.-W., Choi, J.-H., Kim, J.-H., Lee, J.-S.: LarvaNet: hierarchical super-resolution via multi-exit architecture. In: Bartoli, A., Fusiello, A. (eds.) ECCV 2020. LNCS, vol. 12537, pp. 73–86. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-67070-2_4

    Chapter  Google Scholar 

  24. Jiang, Y.G., Cheng, C., Lin, H., Fu, Y.: Learning layer-skippable inference network. IEEE Trans. Image Process. 29, 8747–8759 (2020)

    Article  MATH  Google Scholar 

  25. Jie, Z., Sun, P., Li, X., Feng, J., Liu, W.: Anytime recognition with routing convolutional networks. IEEE Trans. Pattern Anal. Mach. Intell. 43(6), 1875–1886 (2019)

    Article  Google Scholar 

  26. Kaya, Y., Hong, S., Dumitras, T.: Shallow-deep networks: understanding and mitigating network overthinking. In: International Conference on Machine Learning, pp. 3301–3310. PMLR (2019)

    Google Scholar 

  27. Krizhevsky, A., Hinton, G., et al.: Learning multiple layers of features from tiny images (2009)

    Google Scholar 

  28. Li, F., Li, G., He, X., Cheng, J.: Dynamic dual gating neural networks. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 5330–5339 (2021)

    Google Scholar 

  29. Li, H., Zhang, H., Qi, X., Yang, R., Huang, G.: Improved techniques for training adaptive deep networks. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 1891–1900 (2019)

    Google Scholar 

  30. Li, Y., Song, L., Chen, Y., Li, Z., Zhang, X., Wang, X., Sun, J.: Learning dynamic routing for semantic segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 8553–8562 (2020)

    Google Scholar 

  31. Li, Y., Chen, Y.: Revisiting dynamic convolution via matrix decomposition. In: International Conference on Learning Representations (2021)

    Google Scholar 

  32. Liu, B., Rao, Y., Lu, J., Zhou, J., Hsieh, C.-J.: MetaDistiller: network self-boosting via meta-learned top-down distillation. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12359, pp. 694–709. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58568-6_41

    Chapter  Google Scholar 

  33. Liu, B., Liu, X., Jin, X., Stone, P., Liu, Q.: Conflict-averse gradient descent for multi-task learning. In: Advances in Neural Information Processing Systems, vol. 34 (2021)

    Google Scholar 

  34. Liu, L., et al.: Towards impartial multi-task learning. In: ICLR (2021)

    Google Scholar 

  35. Liu, Y., Meng, F., Zhou, J., Chen, Y., Xu, J.: Faster depth-adaptive transformers. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, pp. 13424–13432 (2021)

    Google Scholar 

  36. Ma, N., Zhang, X., Huang, J., Sun, J.: WeightNet: revisiting the design space of weight networks. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12360, pp. 776–792. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58555-6_46

    Chapter  Google Scholar 

  37. Molchanov, P., Tyree, S., Karras, T., Aila, T., Kautz, J.: Pruning convolutional neural networks for resource efficient inference. In: 5th International Conference on Learning Representations, ICLR 2017-Conference Track Proceedings (2019)

    Google Scholar 

  38. Nichol, A., Achiam, J., Schulman, J.: On first-order meta-learning algorithms. arXiv preprint arXiv:1803.02999 (2018)

  39. Passalis, N., Raitoharju, J., Tefas, A., Gabbouj, M.: Efficient adaptive inference for deep convolutional neural networks using hierarchical early exits. Pattern Recogn. 105, 107346 (2020)

    Article  Google Scholar 

  40. Phuong, M., Lampert, C.H.: Distillation-based training for multi-exit architectures. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 1355–1364 (2019)

    Google Scholar 

  41. Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., Chen, L.C.: MobileNetV2: inverted residuals and linear bottlenecks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4510–4520 (2018)

    Google Scholar 

  42. Schwartz, R., Stanovsky, G., Swayamdipta, S., Dodge, J., Smith, N.A.: The right tool for the job: matching model and instance complexities. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pp. 6640–6651 (2020)

    Google Scholar 

  43. Sener, O., Koltun, V.: Multi-task learning as multi-objective optimization. In: Advances in Neural Information Processing Systems, vol. 31 (2018)

    Google Scholar 

  44. Shen, J., Wang, Y., Xu, P., Fu, Y., Wang, Z., Lin, Y.: Fractional skipping: towards finer-grained dynamic CNN inference. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 34, pp. 5700–5708 (2020)

    Google Scholar 

  45. Tan, M., Le, Q.: Efficientnet: rethinking model scaling for convolutional neural networks. In: International Conference on Machine Learning, pp. 6105–6114. PMLR (2019)

    Google Scholar 

  46. Veit, A., Belongie, S.: Convolutional networks with adaptive inference graphs. In: Proceedings of the European Conference on Computer Vision (ECCV), pp. 3–18 (2018)

    Google Scholar 

  47. Wang, X., Yu, F., Dou, Z.Y., Darrell, T., Gonzalez, J.E.: Skipnet: learning dynamic routing in convolutional networks. In: Proceedings of the European Conference on Computer Vision (ECCV), pp. 409–424 (2018)

    Google Scholar 

  48. Wang, X., Li, Y.: Gradient deconfliction-based training for multi-exit architectures. In: 2020 IEEE International Conference on Image Processing (ICIP), pp. 1866–1870. IEEE (2020)

    Google Scholar 

  49. Wang, X., Li, Y.: Harmonized dense knowledge distillation training for multi-exit architectures. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, pp. 10218–10226 (2021)

    Google Scholar 

  50. Wołczyk, M., et al.: Zero time waste: recycling predictions in early exit neural networks. Adv. Neural. Inf. Process. Syst. 34, 2516–2528 (2021)

    Google Scholar 

  51. Wu, Z., et al.: Blockdrop: dynamic inference paths in residual networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 8817–8826 (2018)

    Google Scholar 

  52. Xin, J., Tang, R., Lee, J., Yu, Y., Lin, J.: Deebert: dynamic early exiting for accelerating bert inference. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pp. 2246–2251 (2020)

    Google Scholar 

  53. Yang, B., Bender, G., Le, Q.V., Ngiam, J.: Condconv: conditionally parameterized convolutions for efficient inference. In: Advances in Neural Information Processing Systems, pp. 1307–1318 (2019)

    Google Scholar 

  54. Yang, L., Han, Y., Chen, X., Song, S., Dai, J., Huang, G.: Resolution adaptive networks for efficient inference. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 2369–2378 (2020)

    Google Scholar 

  55. Yu, T., Kumar, S., Gupta, A., Levine, S., Hausman, K., Finn, C.: Gradient surgery for multi-task learning. In: Advances in Neural Information Processing Systems, vol. 33 (2020)

    Google Scholar 

  56. Zhou, J., Jampani, V., Pi, Z., Liu, Q., Yang, M.H.: Decoupled dynamic filter networks. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 6647–6656 (2021)

    Google Scholar 

  57. Zhou, W., Xu, C., Ge, T., McAuley, J., Xu, K., Wei, F.: Bert loses patience: fast and robust inference with early exit. Adv. Neural. Inf. Process. Syst. 33, 18330–18341 (2020)

    Google Scholar 

Download references

Acknowledgement

This work was supported by the National Natural Science Foundation of China (NSFC) under Grants 61825305 and 61973311.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Xin Xu .

Editor information

Editors and Affiliations

1 Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (pdf 101 KB)

Rights and permissions

Reprints and permissions

Copyright information

© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Sun, Y., Li, J., Xu, X. (2022). Meta-GF: Training Dynamic-Depth Neural Networks Harmoniously. In: Avidan, S., Brostow, G., Cissé, M., Farinella, G.M., Hassner, T. (eds) Computer Vision – ECCV 2022. ECCV 2022. Lecture Notes in Computer Science, vol 13671. Springer, Cham. https://doi.org/10.1007/978-3-031-20083-0_41

Download citation

  • DOI: https://doi.org/10.1007/978-3-031-20083-0_41

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-20082-3

  • Online ISBN: 978-3-031-20083-0

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics