Abstract
Counterfactual examples explain a prediction by highlighting changes in an instance that flip the outcome of a classifier. This paper proposes TIGTEC, an efficient and modular method for generating sparse, plausible and diverse counterfactual explanations for textual data. TIGTEC is a text editing heuristic that targets and modifies words with high contribution using local feature importance. A new attention-based local feature importance is proposed. Counterfactual candidates are generated and assessed with a cost function integrating a semantic distance, while the solution space is efficiently explored in a beam search fashion. The conducted experiments show the relevance of TIGTEC in terms of success rate, sparsity, diversity and plausibility. This method can be used in both model-specific or model-agnostic way, which makes it very convenient for generating counterfactual explanations.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
- 2.
- 3.
See the documentation on the publicly available repository: https://github.com/milanbhan/tigtec.
- 4.
References
Akiba, T., Sano, S., Yanase, T., Ohta, T., Koyama, M.: Optuna: a next-generation hyperparameter optimization framework. In: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (2019)
Bahdanau, D., Cho, K., Bengio, Y.: Neural machine translation by jointly learning to align and translate. arXiv:1409 (2014)
Barredo Arrieta, A., et al.: Explainable Artificial Intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI. Inf. Fusion 58, 82–115 (2020)
Bhan, M., Achache, N., Legrand, V., Blangero, A., Chesneau, N.: Evaluating self-attention interpretability through human-grounded experimental protocol. arXiv (2023)
Bibal, A., et al.: Is attention explanation? An introduction to the debate. In: Proceedings of the Association for Computational Linguistics (ACL) (2022)
Dathathri, S., et al.: Plug and play language models: a simple approach to controlled text generation. In: 8th International Conference on Learning Representations, ICLR (2020)
Devlin, J., Chang, M., Lee, K., Toutanova, K.: BERT: pre-training of deep bidirectional transformers for language understanding. In: Proceedings of the Association for Computational Linguistics (ACL) (2019)
Fern, X., Pope, Q.: Text counterfactuals via latent optimization and shapley-guided search. In: Proceedings of Conference on Empirical Methods in Natural Language Processing (EMNLP) (2021)
Guidotti, R.: Counterfactual explanations and how to find them: literature review and benchmarking. Data Mining Knowl. Discov. (2022)
Jelinek, F., Mercer, R.L., Bahl, L.R., Baker, J.K.: Perplexity-a measure of the difficulty of speech recognition tasks. J. Acoust. Soc. Am. 62, 63 (1977)
Laugel, T., Lesot, M.J., Marsala, C., Renard, X., Detyniecki, M.: The dangers of post-hoc interpretability: unjustified counterfactual explanations. In: International Joint Conference on Artificial Intelligence (IJCAI) (2019)
Lewis, M., et al.: BART: denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension. In: Proceedings of the Association for Computational Linguistics (ACL) (2020)
Lundberg, S.M., Lee, S.I.: A unified approach to interpreting model predictions. In: Advances in Neural Information Processing Systems. NeurIPS (2017)
Maas, A.L., Daly, R.E., Pham, P.T., Huang, D., Ng, A.Y., Potts, C.: Learning word vectors for sentiment analysis. In: Proceedings of the Association for Computational Linguistics (ACL) (2011)
Madaan, N., Bedathur, S., Saha, D.: Plug and Play Counterfactual Text Generation for Model Robustness. arXiv (2022)
Mazzine, R., Martens, D.: A framework and benchmarking study for counterfactual generating methods on tabular data. CoRR (2021)
Miller, T.: Explanation in artificial intelligence: insights from the social sciences. Artif. Intell. 267, 1–38 (2019)
Molnar, C.: Interpretable Machine Learning, 2nd edn. (2022). https://christophm.github.io/interpretable-ml-book
Morris, J.X., Lifland, E., Yoo, J.Y., Grigsby, J., Jin, D., Qi, Y.: Textattack: a framework for adversarial attacks, data augmentation, and adversarial training in NLP. In: Proceedings of Conference on Empirical Methods in Natural Language Processing (EMNLP) (2020)
Mothilal, R.K., Sharma, A., Tan, C.: Explaining machine learning classifiers through diverse counterfactual explanations. In: Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency (FAT*) (2020)
Papineni, K., Roukos, S., Ward, T., Zhu, W.J.: Bleu: a method for automatic evaluation of machine translation. In: Proceedings of Association for Computational Linguistics (ACL) (2002)
Poyiadzi, R., Sokol, K., Santos-Rodriguez, R., De Bie, T., Flach, P.: FACE: feasible and actionable counterfactual explanations. In: Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society (AIES) (2020)
Radford, A., Wu, J., Child, R., Luan, D., Amodei, D., Sutskever, I.: Language models are unsupervised multitask learners. OpenAI blog (2019)
Raffel, C., et al.: Exploring the limits of transfer learning with a unified text-to-text transformer. J. Mach. Learn. Res. 21, 5485–5551 (2019)
Reimers, N., Gurevych, I.: Sentence-BERT: sentence embeddings using siamese BERT-networks. In: Proceedings of Empirical Methods in Natural Language Processing (EMNLP) (2019)
Ribeiro, M.T., Singh, S., Guestrin, C.: “Why should I trust you?" Explaining the predictions of any classifier. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (2016)
Ross, A., Marasović, A., Peters, M.: Explaining NLP models via minimal contrastive editing (MiCE). In: Findings of the Association for Computational Linguistics (ACL) (2021)
Russell, C.: Efficient search for diverse coherent explanations. In: Proceedings of the Conference on Fairness, Accountability, and Transparency, pp. 20–28. FAT* (2019)
Sanh, V., Debut, L., Chaumond, J., Wolf, T.: DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter (2020)
Vaswani, A., et al.: Attention is all you need. In: Advances in Neural Information Processing Systems (NeurIPS) (2017)
Wu, T., Ribeiro, M.T., Heer, J., Weld, D.: Polyjuice: Generating counterfactuals for explaining, evaluating, and improving models. In: Proceedings of the Association for Computational Linguistics (ACL) and the Joint Conference on Natural Language Processing (JCNLP) (2021)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Ethics declarations
Ethics Statement
Since the training data for mask language models, Sentence Transformers and classifiers can be biased, there is a risk of generating harmful counterfactual examples. One using TIGTEC to explain the predictions of one’s classifier must be aware of these biases in order to stand back and analyze the produced results. On the other hand, by generating unexpected counterfactual examples, we believe that TIGTEC can be useful in detecting bias in the classifier it seeks to explain. Finally, as any method based on deep learning, this method consumes energy, potentially emitting greenhouse gases. It must be used with caution.
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Bhan, M., Vittaut, JN., Chesneau, N., Lesot, MJ. (2023). TIGTEC: Token Importance Guided TExt Counterfactuals. In: Koutra, D., Plant, C., Gomez Rodriguez, M., Baralis, E., Bonchi, F. (eds) Machine Learning and Knowledge Discovery in Databases: Research Track. ECML PKDD 2023. Lecture Notes in Computer Science(), vol 14171. Springer, Cham. https://doi.org/10.1007/978-3-031-43418-1_30
Download citation
DOI: https://doi.org/10.1007/978-3-031-43418-1_30
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-43417-4
Online ISBN: 978-3-031-43418-1
eBook Packages: Computer ScienceComputer Science (R0)