Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
125 tokens/sec
GPT-4o
47 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

XpertAI: uncovering model strategies for sub-manifolds (2403.07486v1)

Published 12 Mar 2024 in cs.LG

Abstract: In recent years, Explainable AI (XAI) methods have facilitated profound validation and knowledge extraction from ML models. While extensively studied for classification, few XAI solutions have addressed the challenges specific to regression models. In regression, explanations need to be precisely formulated to address specific user queries (e.g.\ distinguishing between Why is the output above 0?' andWhy is the output above 50?'). They should furthermore reflect the model's behavior on the relevant data sub-manifold. In this paper, we introduce XpertAI, a framework that disentangles the prediction strategy into multiple range-specific sub-strategies and allows the formulation of precise queries about the model (the `explanandum') as a linear combination of those sub-strategies. XpertAI is formulated generally to work alongside popular XAI attribution techniques, based on occlusion, gradient integration, or reverse propagation. Qualitative and quantitative results, demonstrate the benefits of our approach.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (44)
  1. An improved data-driven methodology and field-test verification of yaw misalignment calibration on wind turbines. Energy Conversion and Management, 266:115786, 2022.
  2. From attribution maps to human-understandable explanations through concept relevance propagation. Nature Machine Intelligence, 5(9):1006–1019, 2023.
  3. Counterfactual shapley additive explanations. In ACM Conference on Fairness, Accountability, and Transparency, pp.  1054–1070, 2022.
  4. Explainable artificial intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI. Inf. Fusion, 58:82–115, 2020.
  5. On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation. PLoS ONE, 10(7):Art. no. e0130140, 07 2015.
  6. How to explain individual classification decisions. The Journal of Machine Learning Research, 11:1803–1831, 2010.
  7. Decoupling pixel flipping and occlusion strategy for consistent xai benchmarks. arXiv preprint arXiv:2401.06654, 2024.
  8. Disentangled explanations of neural network predictions by finding relevant subspaces. arXiv preprint arXiv:2212.14855, 2022.
  9. Deng, L. The mnist database of handwritten digit images for machine learning research. IEEE Signal Processing Magazine, 29(6):141–142, 2012.
  10. Diffeomorphic counterfactuals with generative models. IEEE Transactions on Pattern Analysis and Machine Intelligence, pp.  1–27, 2023.
  11. Deletion and insertion tests in regression models. Journal of Machine Learning Research, 24(290):1–38, 2023.
  12. The meta-pi network: building distributed knowledge representations for robust multisource pattern recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 14(7):751–769, 1992.
  13. Influence of atmospheric conditions on the power production of utility-scale wind turbines in yaw misalignment. Journal of Renewable and Sustainable Energy, 12(6):Art. no. 063307, 2020.
  14. Condition monitoring systems: a systematic literature review on machine-learning methods improving offshore-wind turbine operational management. International Journal of Sustainable Energy, 40(10):923–946, 2021.
  15. Interpretable mixture of experts. Transactions on Machine Learning Research, 2023.
  16. Adaptive mixtures of local experts. Neural Computation, 3(1):79–87, 1991.
  17. Xrai: Better attributions through regions. In IEEE International Conference on Computer Vision, pp.  4947–4956, 2019.
  18. Interpretability beyond feature attribution: Quantitative testing with concept activation vectors (tcav). In International Conference on Machine Learning, pp.  2668–2677, 2018.
  19. Toward explainable artificial intelligence for precision pathology. Annual Review of Pathology: Mechanisms of Disease, 19:541–570, 2024.
  20. On scientific understanding with artificial intelligence. Nature Reviews Physics, 4(12):761–769, 2022.
  21. An explainable ai framework for robust and transparent data-driven wind turbine power curve models. Energy and AI, 15:100328, 2024.
  22. Toward explainable artificial intelligence for regression models: A methodological perspective. IEEE Signal Processing Magazine, 39(4):40–58, 2022.
  23. Preemptively pruning clever-hans strategies in deep neural networks. Information Fusion, 103:Art. no. 102094, 2024.
  24. A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 30:4765–4774, 2017.
  25. Explaining nonlinear classification decisions with deep taylor decomposition. Pattern Recognition, 65:211–222, 2017.
  26. Operational variables for improving industrial wind turbine yaw misalignment early fault detection capabilities using data-driven techniques. IEEE Transactions on Instrumentation and Measurement, 70:1–8, 2021.
  27. Scada data for wind turbine data-driven condition/performance monitoring: A review on state-of-art, challenges and future trends. Wind Engineering, 47(2):422–441, 2023.
  28. Annealed competition of experts for a segmentation and classification of switching dynamics. Neural Computation, 8(2):340–356, 1996.
  29. Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12:2825–2830, 2011.
  30. ”why should i trust you?”: Explaining the predictions of any classifier. In International Conference on Knowledge Discovery and Data Mining, pp.  1135–1144, 2016.
  31. Evaluating the visualization of what a deep neural network has learned. IEEE Transactions on Neural Networks and Learning Systems, 28(11):2660–2673, 2017.
  32. Explaining deep neural networks and beyond: A review of methods and applications. Proceedings of the IEEE, 109(3):247–278, 2021.
  33. Learning important features through propagating activation differences. In International Conference on Machine Learning, volume 70, pp.  3145–3153, 2017.
  34. Counterfactual explanations for arbitrary regression models. arXiv preprint arXiv:2106.15212, 2021.
  35. Disentangling adversarial robustness and generalization. In IEEE Conference on Computer Vision and Pattern Recognition, pp.  6969–6980, June 2019.
  36. The many shapley values for model explanation. In International Conference on Machine Learning, pp.  9269–9278, 2020.
  37. Axiomatic attribution for deep networks. In International Conference on Machine Learning, volume 70, pp.  3319–3328, 2017.
  38. Visualizing data using t-sne. Journal of Machine Learning Research, 9(86):2579–2605, 2008.
  39. Conditional generative models for counterfactual explanations. arXiv preprint arXiv:2101.10123, 2021.
  40. Multi-dimensional concept discovery (MCD): A unifying framework with completeness guarantees. Transactions on Machine Learning Research, 2023. ISSN 2835-8856.
  41. Explainable ai for time series via virtual inspection layers. Pattern Recognition, pp.  110309, 2024.
  42. Visualizing and understanding convolutional networks. In European Conference on Computer Vision, pp.  818–833, 2014.
  43. Counterfactual explanation for regression via disentanglement in latent space. arXiv preprint arXiv:2311.08228, 2023.
  44. Learning deep features for discriminative localization. In IEEE Conference on Computer Vision and Pattern Recognition, pp.  2921–2929, 2016.
Citations (1)
View on Semantic Scholar

Summary

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

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets