隨著深度神經網路在近年來的快速發展，模型的預測結果往往導向黑盒子決策，這使得決策的可解釋性大幅降低。為了提高決策的透明度，可解釋性AI成為一個重要的研究方向。其中，反事實解釋提供了一種理解模型決策的方式，即在需要相反結果時，特徵需要做出什麼變化。透過這種方式，我們可以觀察並理解決策模型的決策邊界，並在最小改變的情況下達到預期目標。
反事實的生成可以透過多種策略來實現，包括啟發式搜尋策略、基於樹的方法，以及深度學習模型等。這些策略各有其特點，但當反事實與因果理論結合時，可以得到更具可解釋性的結果，已有研究嘗試將因果關係引入深度學習模型中。這些努力都在於提高決策模型的可解釋性，讓我們能更好地理解決策模型的決策方式。此研究提出一種新的架構。我們計算各欄位的因果效應，然後將這些具有因果效應的欄位帶入深度學習網路以生成反事實。這種架構的設計目的是考慮到只有對結果有因果影響的欄位才需要被改變。這種方法不僅使反事實的生成更加精確，進而更有效地達到預期目標，更提高了決策模型的可解釋性。

With the rapid development of deep neural networks in recent years, model predictions often lead to black-box decision-making, significantly reducing the interpretability of decisions. To enhance the transparency of decision-making, explainable AI has become a crucial research direction. Among them, counterfactual explanations provide a way to understand model deci-sions, i.e., what changes need to be made to our conditions when we desire an opposite result. Through this approach, we can observe and understand the decision boundaries of the model and achieve the intended goal with minimal changes.
The generation of counterfactuals can be achieved through various strategies, including heuristic search strategies, tree-based methods, and deep learning models. Each of these strategies has its characteristics, but when counterfactuals are combined with causal theory, we can obtain more interpretable results. Many studies have attempted to introduce causal relationships into deep learning models, all aimed at enhancing the interpretability of the models, enabling us to better understand the decision-making process of the models. In this study, we propose a new model. We calculate the causal effect of each field and then incorporate these causal effects into a deep learning network to generate counterfactuals. The purpose of this framework design is to consider that only fields that have a causal impact on the result need to be changed. This method not only improves the interpretability of our model but also allows us to generate counterfactuals more accurately, thereby more effectively achieving the intended goal.

Abadie, A., & Imbens, G. W. (2016). Matching on the estimated propensity score. Econometrica, 84(2), 781-807.
Ali, S., Abuhmed, T., El-Sappagh, S., Muhammad, K., Alonso-Moral, J. M., Confalonieri, R., . . . Herrera, F. (2023). Explainable Artificial Intelligence (XAI): What we know and what is left to attain Trustworthy Artificial Intelligence. Information Fusion, 99, 101805.
Athey, S., & Wager, S. (2019). Estimating treatment effects with causal forests: An application. Observational studies, 5(2), 37-51.
Becker, B. a. K., Ronny. (1996). UCI Machine Learning Repository. Retrieved from https://archive.ics.uci.edu/ml/datasets/adult. from UCI Machine Learning Repository https://archive.ics.uci.edu/ml/datasets/adult
Bodria, F., Guidotti, R., Giannotti, F., & Pedreschi, D. (2022). Transparent Latent Space Counterfactual Explanations for Tabular Data. Paper presented at the 2022 IEEE 9th International Conference on Data Science and Advanced Analytics (DSAA).
Brughmans, D., Leyman, P., & Martens, D. (2023). Nice: an algorithm for nearest instance counterfactual explanations. Data mining and knowledge discovery, 1-39.
Bueff, A. C., Cytryński, M., Calabrese, R., Jones, M., Roberts, J., Moore, J., & Brown, I. (2022). Machine learning interpretability for a stress scenario generation in credit scoring based on counterfactuals. Expert Systems with Applications, 202, 117271.
Caliendo, M., & Kopeinig, S. (2008). Some practical guidance for the implementation of propensity score matching. Journal of economic surveys, 22(1), 31-72.
Chipman, H. A., George, E. I., & McCulloch, R. E. (2010). BART: BAYESIAN ADDITIVE REGRESSION TREES. The Annals of Applied Statistics, 4(1), 266-298. Retrieved from http://www.jstor.org/stable/27801587
Ellis, B., & Wong, W. H. (2008). Learning causal Bayesian network structures from experimental data. Journal of the American Statistical Association, 103(482), 778-789.
Gilmer, J., Schoenholz, S. S., Riley, P. F., Vinyals, O., & Dahl, G. E. (2017). Neural Message Passing for Quantum Chemistry. Paper presented at the Proceedings of the 34th International Conference on Machine Learning, Proceedings of Machine Learning Research. https://proceedings.mlr.press/v70/gilmer17a.html
Guidotti, R. (2022). Counterfactual explanations and how to find them: literature review and benchmarking. Data Mining and Knowledge Discovery. doi:10.1007/s10618-022-00831-6
Gunning, D., & Aha, D. (2019). DARPA’s explainable artificial intelligence (XAI) program. AI magazine, 40(2), 44-58.
Guo, R., Cheng, L., Li, J., Hahn, P. R., & Liu, H. (2020). A survey of learning causality with data: Problems and methods. ACM Computing Surveys (CSUR), 53(4), 1-37.
Kaur, D., Uslu, S., Rittichier, K. J., & Durresi, A. (2022). Trustworthy artificial intelligence: a review. ACM Computing Surveys (CSUR), 55(2), 1-38.
Lucic, A., Haned, H., & de Rijke, M. (2020). Why does my model fail? contrastive local explanations for retail forecasting. Paper presented at the Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency.
Mahajan, D., Tan, C., & Sharma, A. (2019). Preserving causal constraints in counterfactual explanations for machine learning classifiers. arXiv preprint arXiv:1912.03277.
Marchezini, G. F., Lacerda, A. M., Pappa, G. L., Meira Jr, W., Miranda, D., Romano-Silva, M. A., . . . Diniz, L. M. (2022). Counterfactual inference with latent variable and its application in mental health care. Data Mining and Knowledge Discovery, 36(2), 811-840.
Molnar, C. (2019). Interpretable Machine Learning.
Mothilal, R. K., Sharma, A., & Tan, C. (2020). Explaining machine learning classifiers through diverse counterfactual explanations. Paper presented at the Proceedings of the 2020 conference on fairness, accountability, and transparency.
Nemirovsky, D., Thiebaut, N., Xu, Y., & Gupta, A. (2020). Countergan: Generating realistic counterfactuals with residual generative adversarial nets. arXiv preprint arXiv:2009.05199.
Paramasivan, K., Subburaj, R., Jaiswal, S., & Sudarsanam, N. (2022). Empirical evidence of the impact of mobility on property crimes during the first two waves of the COVID-19 pandemic. Humanities and Social Sciences Communications, 9(1), 1-14.
Pawelczyk, M., Broelemann, K., & Kasneci, G. (2020). Learning model-agnostic counterfactual explanations for tabular data. Paper presented at the Proceedings of the web conference 2020.
Pearl, J. (2000). Models, reasoning and inference. Cambridge, UK: CambridgeUniversityPress, 19(2), 3.
Rosenbaum, P. R., & Rubin, D. B. (1983). The central role of the propensity score in observational studies for causal effects. Biometrika, 70(1), 41-55.
Schwab, P., Linhardt, L., Bauer, S., Buhmann, J. M., & Karlen, W. (2020). Learning counterfactual representations for estimating individual dose-response curves. Paper presented at the Proceedings of the AAAI Conference on Artificial Intelligence.
Seaman, S. R., & White, I. R. (2013). Review of inverse probability weighting for dealing with missing data. Statistical methods in medical research, 22(3), 278-295.
Shalit, U., Johansson, F. D., & Sontag, D. (2017). Estimating individual treatment effect: generalization bounds and algorithms. Paper presented at the International conference on machine learning.
Shi, C., Blei, D. M., & Veitch, V. (2019). Adapting neural networks for the estimation of treatment effects. In Proceedings of the 33rd International Conference on Neural Information Processing Systems (pp. Article 225): Curran Associates Inc.
Sohn, K., Lee, H., & Yan, X. (2015). Learning structured output representation using deep conditional generative models. Advances in neural information processing systems, 28.
Stepin, I., Alonso, J. M., Catala, A., & Pereira-Fariña, M. (2021). A survey of contrastive and counterfactual explanation generation methods for explainable artificial intelligence. IEEE Access, 9, 11974-12001.
Van Looveren, A., & Klaise, J. (2021). Interpretable counterfactual explanations guided by prototypes. Paper presented at the Joint European Conference on Machine Learning and Knowledge Discovery in Databases.
Wachter, S., Mittelstadt, B., & Russell, C. (2017). Counterfactual explanations without opening the black box: Automated decisions and the GDPR. Harv. JL & Tech., 31, 841.
Wachter, S., Mittelstadt, B. D., & Russell, C. (2017). Counterfactual Explanations Without Opening the Black Box: Automated Decisions and the GDPR. Cybersecurity.
Xiang, X., & Lenskiy, A. (2022). Realistic Counterfactual Explanations with Learned Relations. arXiv preprint arXiv:2202.07356.
Xu, F., Uszkoreit, H., Du, Y., Fan, W., Zhao, D., & Zhu, J. (2019). Explainable AI: A brief survey on history, research areas, approaches and challenges. Paper presented at the Natural Language Processing and Chinese Computing: 8th CCF International Conference, NLPCC 2019, Dunhuang, China, October 9–14, 2019, Proceedings, Part II 8.
Xu, R., Yu, Y., Zhang, C., Ali, M. K., Ho, J. C., & Yang, C. (2022). Counterfactual and Factual Reasoning over Hypergraphs for Interpretable Clinical Predictions on EHR. Paper presented at the Proceedings of the 2nd Machine Learning for Health symposium, Proceedings of Machine Learning Research. https://proceedings.mlr.press/v193/xu22a.html
Yang, M., Liu, F., Chen, Z., Shen, X., Hao, J., & Wang, J. (2020). Causalvae: Structured causal disentanglement in variational autoencoder. arXiv preprint arXiv:2004.08697.
Yao, L., Chu, Z., Li, S., Li, Y., Gao, J., & Zhang, A. (2021). A survey on causal inference. ACM Transactions on Knowledge Discovery from Data (TKDD), 15(5), 1-46.
Yao, L., Li, S., Li, Y., Huai, M., Gao, J., & Zhang, A. (2018). Representation learning for treatment effect estimation from observational data. Advances in neural information processing systems, 31.
Zhang, X., Dai, L., Peng, Q., Tang, R., & Li, X. (2023). A Survey of Counterfactual Explanations: Definition, Evaluation, Algorithms, and Applications. In (pp. 905-912).
Zhou, S., Islam, U. J., Pfeiffer, N., Banerjee, I., Patel, B. K., & Iquebal, A. S. (2023). SCGAN: Sparse CounterGAN for Counterfactual Explanations in Breast Cancer Prediction. IEEE Transactions on Automation Science and Engineering.