Skip to main content
Springer Nature Link
Log in

\( Xpression \): A Unifying Metric to Optimize Compression and Explainability Robustness of AI Models

  • Conference paper
  • First Online:
Explainable Artificial Intelligence (xAI 2024)

Part of the book series: Communications in Computer and Information Science ((CCIS,volume 2153))

Included in the following conference series:

  • 854 Accesses

Abstract

Trustworthiness and efficiency have recently become crucial aspects of applied AI. The intersection of interpretability and model compression, however, still poses significant challenges. As models undergo compression for improved efficiency, maintaining explainability needs to remain a priority. In this paper, we propose a novel metric to evaluate both aspects simultaneously and help practitioners navigate this trade-off. In particular, we delve into the effect that knowledge distillation, quantization, and pruning have on the Infidelity explainability metric. Our goal is for \( Xpression \) metric to guide the optimization of compression whilst the model keeps its infidelity robustness. Experimental results across several neural network architectures show the effectiveness of the proposed metric in combining efficiency and relative interpretability with respect to the original model. This work contributes to advancing the understanding of compression techniques and provides a valuable framework for evaluating their implications on model interpretability.

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

Access this chapter

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

References

  1. Xie, Q., Luong, M.-T., Hovy, E., Le, Q.V.: Self-training with noisy student improves ImageNet classification. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2020)

    Google Scholar 

  2. Zhu, M., Gupta, S.: To prune, or not to prune: exploring the efficacy of pruning for model compression (2017). arXiv:1710.01878

  3. Wu, H., Judd, P., Zhang, X., Isaev, M., Micikevicius, P.: Integer quantization for deep learning inference: Principles and empirical evaluation (2020). arXiv preprint arXiv:2004.09602

  4. Gou, J., Yu, B., Maybank, S.J., Tao, D.: Knowledge distillation: a survey. Int. J. Comput. Vis. (IJCV) 129(6), 1789–1819 (2021)

    Google Scholar 

  5. Choudhary, T., Mishra, V., Goswami, A., Sarangapani, J.: A comprehensive survey on model compression and acceleration. Artif. Intell. Rev. 53, 5113–5155 (2020)

    Google Scholar 

  6. Bell, A., Solano-Kamaiko, I., Nov, O., Stoyanovich, J.: It’s just not that simple: an empirical study of the accuracy-explainability trade-off in machine learning for public policy. In: ACM Conference on Fairness, Accountability, and Transparency, pp. 248–266 (2022)

    Google Scholar 

  7. Wu, C.-J., et al.: Sustainable AI: Environmental implications, challenges and opportunities (2021). ArXiv, abs/2111.00364

    Google Scholar 

  8. Batic, D., Tanoni, G., Stankovic, L., Stankovic, V., Principi, E.: Improving knowledge distillation for non-intrusive load monitoring through explainability guided learning. In: Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 1–5. IEEE (2023)

    Google Scholar 

  9. Alharbi, R., Vu, M.N., Thai, M.T.: Learning interpretation with explainable knowledge distillation. In: Proceedings of the 2021 IEEE International Conference on Big Data (Big Data), pp. 705–714 (2021)

    Google Scholar 

  10. Alharbi, R., Vu, M.N., Thai, M.T.: Dissecting pruned neural networks. In: International Conference on Learning Representations workshop (ICLRw) (2019)

    Google Scholar 

  11. Luo, X., Chi, W., Deng, M.: Deepprune: Learning efficient and interpretable convolutional networks through weight pruning for predicting DNA-protein binding. Front. Genet. 10, 1145 (2019)

    Google Scholar 

  12. Becking, D., Dreyer, M., Samek, W., Müller, K., Lapuschkin, S.: ECQx: explainability-driven quantization for low-bit and sparse DNNs. In: International Workshop on Extending Explainable AI Beyond Deep Models and Classifiers (2020)

    Google Scholar 

  13. Tashiro, Y., Awano, H.: Pay attention via quantization: enhancing explainability of neural networks via quantized activation. IEEE Access 11, 34431–34439 (2023)

    Google Scholar 

  14. Dardouillet, P., Benoit, A., Amri, E., Bolon, P., Dubucq, D., Crédoz, A.: Explainability of image semantic segmentation through SHAP values. In: Proceedings of the International Conference on Pattern Recognition (ICPR), pp. 188–202 (2022)

    Google Scholar 

  15. Karri, M., Annavarapu, C.S.R., Rajendra Acharya, U.: Explainable multi-module semantic guided attention based network for medical image segmentation. Comput. Biol. Med. 151, 106231 (2022)

    Google Scholar 

  16. Yeh, C.-K., Hsieh, C.-Y., Suggala, A., Inouye, D.I., Ravikumar, P.K.: On the (in) fidelity and sensitivity of explanations. In: Advances in Neural Information Processing Systems, vol. 32 (2019)

    Google Scholar 

  17. Longo, L., et al.: Explainable artificial intelligence (XAI) 2.0: a manifesto of open challenges and interdisciplinary research directions. Inf. Fusion 106, 102301 (2024)

    Google Scholar 

  18. Hoefler, T., Alistarh, D., Ben-Nun, T., Dryden, N., Peste, A.: Sparsity in deep learning: pruning and growth for efficient inference and training in neural networks. J. Mach. Learn. Res. 22(241), 1–124 (2021)

    MathSciNet  Google Scholar 

  19. Hassija, V., et al.: Interpreting black-box models: a review on explainable artificial intelligence. Cogn. Comput. 16, 45–74 (2024)

    Article  Google Scholar 

  20. Mishra, S., Dutta, S., Long, J., Magazzeni, D.: A survey on the robustness of feature importance and counterfactual explanations. In: Workshop on Explainable AI in Finance (XAI-FIN21) (2021)

    Google Scholar 

  21. Selvaraju, R.R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., Batra, D.: Grad-CAM: visual explanations from deep networks via gradient-based localization. In: Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp. 618–626 (2017)

    Google Scholar 

  22. Simonyan, K., Vedaldi, A., Zisserman, A.: Deep inside convolutional networks: Visualising image classification models and saliency maps (2013). arXiv preprint arXiv:1312.6034

  23. Lundberg, S.M., Lee, S.-I.: A unified approach to interpreting model predictions. In: Advances in Neural Information Processing Systems NeurIPS, vol. 30 (2017)

    Google Scholar 

  24. Ribeiro, M.T., Singh, S., Guestrin, C.: why should i trust you? explaining the predictions of any classifier. In: ACM International Conference on Knowledge Discovery and Data Mining (SIGKDD), pp. 1135–1144 (2016)

    Google Scholar 

  25. Ba, J., Caruana, R.: Do deep nets really need to be deep? In: Advances in Neural Information Processing Systems (NeurIPS), vol. 27 (2014)

    Google Scholar 

  26. Hinton, G., Vinyals, O., Dean, J.: Distilling the knowledge in a neural network. Advances in Neural Information Processing Systems (NeurIPS), vol. 27 (2014)

    Google Scholar 

  27. Zagoruyko, S., Komodakis, N.: Paying more attention to attention: improving the performance of convolutional neural networks via attention transfer. In: International Conference on Learning Representations (ICLR) (2017)

    Google Scholar 

  28. Lee, S.H., Kim, D.H., Song, B.C.: Self-supervised knowledge distillation using singular value decomposition. In: Proceedings of the European Conference on Computer Vision (ECCV) (2018)

    Google Scholar 

  29. Tung, F., Mori, G.: Similarity-preserving knowledge distillation. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (CVPR), pp. 335–350 (2019)

    Google Scholar 

  30. Zhang, Y., Xiang, T., Hospedales, T.M., Lu, H.: Deep mutual learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4320–4328 (2018)

    Google Scholar 

  31. Chung, I., Park, S., Kim, J., Kwak, N.: Feature-map-level online adversarial knowledge distillation. In: Proceedings of the International Conference on Machine Learning (ICML), pp. 2006–2015 (2020)

    Google Scholar 

  32. Tian, Y., Krishnan, D., Isola, P.: Contrastive representation distillation. In: International Conference on Learning Representations (ICLR) (2020)

    Google Scholar 

  33. Frankle, J., Carbin, M.: The lottery ticket hypothesis: finding sparse, trainable neural networks. In: International Conference on Learning Representations (ICLR) (2019)

    Google Scholar 

  34. Li, H., Kadav, A., Durdanovic, I., Samet, H., Graf, H.P.: Pruning filters for efficient convnets. In: International Conference on Learning Representations (ICLR) (2018)

    Google Scholar 

  35. Liu, Z., Li, J., Shen, Z., Huang, G., Yan, S., Zhang, C.: Learning efficient convolutional networks through network slimming. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pp. 2736–2744 (2017)

    Google Scholar 

  36. Yu, R., et al.: NISP: pruning networks using neuron importance score propagation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 9194–9203 (2018)

    Google Scholar 

  37. Wang, H., Qin, C., Zhang, Y., Fu, Y.: Neural pruning via growing regularization. In: International Conference on Learning Representations (ICLR) (2020)

    Google Scholar 

  38. Fang, G., Ma, X., Song, M., Mi, M.B., Wang, X.: DepGraph: towards any structural pruning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 16091–16101 (2023)

    Google Scholar 

  39. Gholami, A., Kim, S., Dong, Z., Yao, Z., Mahoney, M.W., Keutzer, K.: A survey of quantization methods for efficient neural network inference. In: Low-Power Computer Vision. Chapman and Hall/CRC (2022)

    Google Scholar 

  40. Gray, R.M., Neuhoff, D.L.: Quantization. IEEE Trans. Inf. Theor. 44, 2325–2383 (1998)

    Article  Google Scholar 

  41. Han, S., Mao, H., Dally, W.J.: Deep compression: compressing deep neural networks with pruning, trained quantization and huffman coding. In: International Conference on Learning Representations (ICLR) (2015)

    Google Scholar 

  42. Nagel, M., Amjad, R.A., Van Baalen, M., Louizos, C., Blankevoort, T.: Up or down? adaptive rounding for post-training quantization. In: Proceedings of the International Conference on Machine Learning (ICML), pp. 7197–7206 (2020)

    Google Scholar 

  43. Zhewei Yao, et al.: HAWQ-V3: dyadic neural network quantization. In: Proceedings of the International Conference on Machine Learning (ICML), pp. 11875–11886 (2021)

    Google Scholar 

  44. Dong, Z., Yao, Z., Arfeen, D., Gholami, A., Mahoney, M.W., Keutzer, K.: HAWQ-V2: hessian aware trace-weighted quantization of neural networks. In: Advances in Neural Information Processing Systems (NeurIPS), pp. 33 (2020)

    Google Scholar 

  45. Dong, Z., Yao, Z., Gholami, A., Mahoney, M.W., Keutzer, K.: Hawq: Hessian aware quantization of neural networks with mixed-precision. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), vol. 32 (2019)

    Google Scholar 

  46. Yang, J., et al.: Quantization networks. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 12434–12443 (2019)

    Google Scholar 

  47. Wang, L., Dong, X., Wang, Y., Liu, L., An, W., Guo, Y.: Learnable lookup table for neural network quantization. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 12423–12433 (2022)

    Google Scholar 

  48. Ma, Y., et al.: OMPQ: orthogonal mixed precision quantization. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 37, pp. 9029–9037 (2023)

    Google Scholar 

  49. Sun, T., Chen, H., Hu, G., Zhao, C.: Explainability-based knowledge distillation (2023). Available at SSRN 4460609

    Google Scholar 

  50. Sousa, J., Moreira, R., Balayan, V., Saleiro, P., Bizarro, P.: ConceptDistil: model-agnostic distillation of concept explanations. In: International Conference on Learning Representations (ICLR) (2022)

    Google Scholar 

  51. Termritthikun, C., Umer, A., Suwanwimolkul, S., Xia, F., Lee, I.: Explainable knowledge distillation for on-device chest X-ray classification. IEEE/ACM Trans. Comput. Biol. Bioinform. 1–12 (2023)

    Google Scholar 

  52. Liu, X., Wang, X., Matwin, S.: Improving the interpretability of deep neural networks with knowledge distillation. In: IEEE International Conference on Data Mining Workshops (ICDMW) (2018)

    Google Scholar 

  53. Li, Y., Liu, L., Wang, G., Yong, D., Chen, P.: EGNN: constructing explainable graph neural networks via knowledge distillation. Knowl. Based Syst. 241, 108345 (2022)

    Article  Google Scholar 

  54. Han, H., Kim, S., Choi, H.-S., Yoon, S.: On the impact of knowledge distillation for model interpretability. In: Proceedings of the International Conference on Machine Learning (ICML), pp. 12389–12410 (2023)

    Google Scholar 

  55. Weber, D., Merkle, F., Schöttle, P., Schlögl, S.: Less is more: The influence of pruning on the explainability of CNNs (2023). arXiv:2302.08878

  56. Norrenbrock, T., Rudolph, M., Rosenhahn, B.: Q-SENN: quantized self-explaining neural networks. In: Proceedings of the AAAI Conference on Artificial Intelligence, pp. 21482–21491 (2023)

    Google Scholar 

  57. Sabih, M., Hannig, F., Teich, J.: Utilizing explainable AI for quantization and pruning of deep neural networks (2020). arXiv:2008.09072

  58. Smilkov, D., Kim, B., Thorat, N., Viégas, F., Wattenberg, M.: SmoothGrad: removing noise by adding noise (2017). arXiv:1706.03825

  59. Krizhevsky, A., Hinton, G.: Learning multiple layers of features from tiny images. Technical report, University of Toronto (2009)

    Google Scholar 

  60. 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 (CVPR), pp. 770–778 (2016)

    Google Scholar 

  61. Hu, Y.: Knowledge distillation zoo. GitHub Repos. (2019). GitHub. https://github.com/AberHu/Knowledge-Distillation-Zoo

  62. Li, Y., Dong, X., Wang, W.: Additive powers-of-two quantization: an efficient non-uniform discretization for neural networks. In: International Conference on Learning Representations (ICLR) (2020)

    Google Scholar 

  63. Kokhlikyan, N., et al.: A Unified and Generic Model Interpretability Library for Pytorch, Captum (2020)

    Google Scholar 

Download references

Acknowledgments

The authors want to thank the European Commission for the funding under the Horizon Europe programme MANOLO Grant Agreement No.101135782.

Author information

Authors and Affiliations

  1. CeADAR - Ireland Centre for Applied AI, University College Dublin, Dublin, Ireland

    Eric Arazo, Hristo Stoev, Cristian Bosch, Andrés L. Suárez-Cetrulo & Ricardo Simón-Carbajo

Authors
  1. Eric Arazo
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Hristo Stoev
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Cristian Bosch
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Andrés L. Suárez-Cetrulo
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Ricardo Simón-Carbajo
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Eric Arazo .

Editor information

Editors and Affiliations

  1. Technological University Dublin, Dublin, Ireland

    Luca Longo

  2. Fraunhofer Institute for Telecommunications, Heinrich-Hertz-Institut, HHI, Berlin, Germany

    Sebastian Lapuschkin

  3. University of Marburg, Marburg, Germany

    Christin Seifert

Ethics declarations

Disclosure of Interests

The authors have no competing interests to declare that are relevant to the content of this article.

Rights and permissions

Reprints and permissions

Copyright information

© 2024 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

Arazo, E., Stoev, H., Bosch, C., Suárez-Cetrulo, A.L., Simón-Carbajo, R. (2024). \( Xpression \): A Unifying Metric to Optimize Compression and Explainability Robustness of AI Models. In: Longo, L., Lapuschkin, S., Seifert, C. (eds) Explainable Artificial Intelligence. xAI 2024. Communications in Computer and Information Science, vol 2153. Springer, Cham. https://doi.org/10.1007/978-3-031-63787-2_19

Download citation

  • DOI: https://doi.org/10.1007/978-3-031-63787-2_19

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-63786-5

  • Online ISBN: 978-3-031-63787-2

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics