{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,27]],"date-time":"2026-09-27T12:04:35Z","timestamp":1790510675820,"version":"4.1.0"},"reference-count":43,"publisher":"Association for Computing Machinery (ACM)","issue":"1","license":[{"start":{"date-parts":[[2021,5,26]],"date-time":"2021-05-26T00:00:00Z","timestamp":1621987200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["SIGKDD Explor. Newsl."],"published-print":{"date-parts":[[2021,5,26]]},"abstract":"<jats:p>With the wide use of deep neural networks (DNN), model interpretability has become a critical concern, since explainable decisions are preferred in high-stake scenarios. Current interpretation techniques mainly focus on the feature attribution perspective, which are limited in indicating why and how particular explanations are related to the prediction. To this end, an intriguing class of explanations, named counterfactuals, has been developed to further explore the \"what-if\" circumstances for interpretation, and enables the reasoning capability on black-box models. However, generating counterfactuals for raw data instances (i.e., text and image) is still in the early stage due to its challenges on high data dimensionality and unsemantic raw features. In this paper, we design a framework to generate counterfactuals specifically for raw data instances with the proposed Attribute-Informed Perturbation (AIP). By utilizing generative models conditioned with different attributes, counterfactuals with desired labels can be obtained effectively and efficiently. Instead of directly modifying instances in the data space, we iteratively optimize the constructed attributeinformed latent space, where features are more robust and semantic. Experimental results on real-world texts and images demonstrate the effectiveness, sample quality as well as efficiency of our designed framework, and show the superiority over other alternatives. Besides, we also introduce some practical applications based on our framework, indicating its potential beyond the model interpretability aspect.<\/jats:p>","DOI":"10.1145\/3468507.3468517","type":"journal-article","created":{"date-parts":[[2021,5,30]],"date-time":"2021-05-30T00:55:35Z","timestamp":1622336135000},"page":"59-68","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":13,"title":["Generative Counterfactuals for Neural Networks via Attribute-Informed Perturbation"],"prefix":"10.1145","volume":"23","author":[{"given":"Fan","family":"Yang","sequence":"first","affiliation":[{"name":"Texas A&amp;M University, College Station, TX, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ninghao","family":"Liu","sequence":"additional","affiliation":[{"name":"Texas A&amp;M University, College Station, TX, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mengnan","family":"Du","sequence":"additional","affiliation":[{"name":"Texas A&amp;M University, College Station, TX, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xia","family":"Hu","sequence":"additional","affiliation":[{"name":"Texas A&amp;M University, College Station, TX, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2021,5,29]]},"reference":[{"key":"e_1_2_1_1_1","volume-title":"Removing input features via a generative model to explain their attributions to classifier's decisions. arXiv preprint arXiv:1910.04256","author":"Agarwal C.","year":"2019","unstructured":"C. Agarwal , D. Schonfeld , and A. Nguyen . Removing input features via a generative model to explain their attributions to classifier's decisions. arXiv preprint arXiv:1910.04256 , 2019 . C. Agarwal, D. Schonfeld, and A. Nguyen. Removing input features via a generative model to explain their attributions to classifier's decisions. arXiv preprint arXiv:1910.04256, 2019."},{"key":"e_1_2_1_2_1","volume-title":"Yelp dataset challenge: Review rating prediction. arXiv preprint arXiv:1605.05362","author":"Asghar N.","year":"2016","unstructured":"N. Asghar . Yelp dataset challenge: Review rating prediction. arXiv preprint arXiv:1605.05362 , 2016 . N. Asghar. Yelp dataset challenge: Review rating prediction. arXiv preprint arXiv:1605.05362, 2016."},{"key":"e_1_2_1_3_1","first-page":"8928","volume-title":"Advances in Neural Information Processing Systems","author":"Chen C.","year":"2019","unstructured":"C. Chen , O. Li , D. Tao , A. Barnett , C. Rudin , and J. K. Su . This looks like that: deep learning for interpretable image recognition . In Advances in Neural Information Processing Systems , pages 8928 -- 8939 , 2019 . C. Chen, O. Li, D. Tao, A. Barnett, C. Rudin, and J. K. Su. This looks like that: deep learning for interpretable image recognition. In Advances in Neural Information Processing Systems, pages 8928--8939, 2019."},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/E17-1104"},{"key":"e_1_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.5555\/3326943.3326998"},{"key":"e_1_2_1_6_1","volume-title":"Towards a rigorous science of interpretable machine learning. arXiv preprint arXiv:1702.08608","author":"Doshi-Velez F.","year":"2017","unstructured":"F. Doshi-Velez and B. Kim . Towards a rigorous science of interpretable machine learning. arXiv preprint arXiv:1702.08608 , 2017 . F. Doshi-Velez and B. Kim. Towards a rigorous science of interpretable machine learning. arXiv preprint arXiv:1702.08608, 2017."},{"key":"e_1_2_1_7_1","doi-asserted-by":"publisher","DOI":"10.1145\/3359786"},{"key":"e_1_2_1_8_1","volume-title":"Cdeepex: Contrastive deep explanations","author":"Feghahati A.","year":"2018","unstructured":"A. Feghahati , C. R. Shelton , M. J. Pazzani , and K. Tang . Cdeepex: Contrastive deep explanations . 2018 . A. Feghahati, C. R. Shelton, M. J. Pazzani, and K. Tang. Cdeepex: Contrastive deep explanations. 2018."},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.1109\/SPW.2018.00016"},{"key":"e_1_2_1_10_1","doi-asserted-by":"publisher","DOI":"10.5555\/2969033.2969125"},{"key":"e_1_2_1_11_1","volume-title":"3rd International Conference on Learning Representations","author":"Goodfellow I. J.","year":"2015","unstructured":"I. J. Goodfellow , J. Shlens , and C. Szegedy . Explaining and harnessing adversarial examples . In 3rd International Conference on Learning Representations , 2015 . I. J. Goodfellow, J. Shlens, and C. Szegedy. Explaining and harnessing adversarial examples. In 3rd International Conference on Learning Representations, 2015."},{"key":"e_1_2_1_12_1","first-page":"2376","volume-title":"ICML","author":"Goyal Y.","year":"2019","unstructured":"Y. Goyal , Z. Wu , J. Ernst , D. Batra , D. Parikh , and S. Lee . Counterfactual visual explanations . In ICML , pages 2376 -- 2384 , 2019 . Y. Goyal, Z. Wu, J. Ernst, D. Batra, D. Parikh, and S. Lee. Counterfactual visual explanations. In ICML, pages 2376--2384, 2019."},{"key":"e_1_2_1_13_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICBDA.2017.8078730"},{"key":"e_1_2_1_14_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.1145\/2872427.2883037"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1109\/tip.2019.2916751"},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.5555\/3305381.3305545"},{"key":"e_1_2_1_18_1","volume-title":"xgems: Generating examplars to explain black-box models. arXiv preprint arXiv:1806.08867","author":"Joshi S.","year":"2018","unstructured":"S. Joshi , O. Koyejo , B. Kim , and J. Ghosh . xgems: Generating examplars to explain black-box models. arXiv preprint arXiv:1806.08867 , 2018 . S. Joshi, O. Koyejo, B. Kim, and J. Ghosh. xgems: Generating examplars to explain black-box models. arXiv preprint arXiv:1806.08867, 2018."},{"key":"e_1_2_1_19_1","doi-asserted-by":"publisher","DOI":"10.5555\/3157096.3157352"},{"key":"e_1_2_1_20_1","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/D14-1181"},{"key":"e_1_2_1_21_1","volume-title":"Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114","author":"Kingma D. P.","year":"2013","unstructured":"D. P. Kingma and M. Welling . Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114 , 2013 . D. P. Kingma and M. Welling. Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114, 2013."},{"key":"e_1_2_1_22_1","volume-title":"Textbugger: Generating adversarial text against real-world applications. arXiv preprint arXiv:1812.05271","author":"Li J.","year":"2018","unstructured":"J. Li , S. Ji , T. Du , B. Li , and T. Wang . Textbugger: Generating adversarial text against real-world applications. arXiv preprint arXiv:1812.05271 , 2018 . J. Li, S. Ji, T. Du, B. Li, and T. Wang. Textbugger: Generating adversarial text against real-world applications. arXiv preprint arXiv:1812.05271, 2018."},{"key":"e_1_2_1_23_1","volume-title":"Lawrence Livermore National Lab.(LLNL)","author":"Liu S.","year":"2019","unstructured":"S. Liu , B. Kailkhura , D. Loveland , and H. Yong . Generative counterfactual introspection for explainable deep learning. Technical report , Lawrence Livermore National Lab.(LLNL) , 2019 . S. Liu, B. Kailkhura, D. Loveland, and H. Yong. Generative counterfactual introspection for explainable deep learning. Technical report, Lawrence Livermore National Lab.(LLNL), 2019."},{"key":"e_1_2_1_24_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.425"},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.5555\/3295222.3295230"},{"key":"e_1_2_1_26_1","volume-title":"A structural-model approach. arXiv preprint arXiv:1811.03163","author":"Miller T.","year":"2018","unstructured":"T. Miller . Contrastive explanation : A structural-model approach. arXiv preprint arXiv:1811.03163 , 2018 . T. Miller. Contrastive explanation: A structural-model approach. arXiv preprint arXiv:1811.03163, 2018."},{"key":"e_1_2_1_27_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-29908-8_4"},{"key":"e_1_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.5555\/2380985"},{"key":"e_1_2_1_29_1","doi-asserted-by":"publisher","DOI":"10.1214\/09-SS057"},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.5555\/3238230"},{"key":"e_1_2_1_31_1","doi-asserted-by":"publisher","DOI":"10.1145\/3234150"},{"key":"e_1_2_1_32_1","doi-asserted-by":"publisher","DOI":"10.1145\/2939672.2939778"},{"key":"e_1_2_1_33_1","volume-title":"32nd AAAI Conference on Artificial Intelligence","author":"Ribeiro M. T.","year":"2018","unstructured":"M. T. Ribeiro , S. Singh , and C. Guestrin . Anchors: High-precision model-agnostic explanations . In 32nd AAAI Conference on Artificial Intelligence , 2018 . M. T. Ribeiro, S. Singh, and C. Guestrin. Anchors: High-precision model-agnostic explanations. In 32nd AAAI Conference on Artificial Intelligence, 2018."},{"key":"e_1_2_1_34_1","first-page":"187","volume-title":"Example-based reasoning. Informal reasoning in education","author":"Rissland E. L.","year":"1991","unstructured":"E. L. Rissland . Example-based reasoning. Informal reasoning in education , pages 187 -- 208 , 1991 . E. L. Rissland. Example-based reasoning. Informal reasoning in education, pages 187--208, 1991."},{"key":"e_1_2_1_35_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01249-6_41"},{"key":"e_1_2_1_36_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.74"},{"key":"e_1_2_1_37_1","volume-title":"Should adversarial attacks use pixel p-norm? arXiv preprint arXiv:1906.02439","author":"Sen A.","year":"2019","unstructured":"A. Sen , X. Zhu , L. Marshall , and R. Nowak . Should adversarial attacks use pixel p-norm? arXiv preprint arXiv:1906.02439 , 2019 . A. Sen, X. Zhu, L. Marshall, and R. Nowak. Should adversarial attacks use pixel p-norm? arXiv preprint arXiv:1906.02439, 2019."},{"key":"e_1_2_1_38_1","volume-title":"8th International Conference on Learning Representations","author":"Singla S.","year":"2020","unstructured":"S. Singla , B. Pollack , J. Chen , and K. Batmanghelich . Explanation by progressive exaggeration . In 8th International Conference on Learning Representations , 2020 . S. Singla, B. Pollack, J. Chen, and K. Batmanghelich. Explanation by progressive exaggeration. In 8th International Conference on Learning Representations, 2020."},{"key":"e_1_2_1_39_1","volume-title":"Smoothgrad: removing noise by adding noise. arXiv preprint arXiv:1706.03825","author":"Smilkov D.","year":"2017","unstructured":"D. Smilkov , N. Thorat , B. Kim , F. Vi\u00b4egas , and M. Wattenberg . Smoothgrad: removing noise by adding noise. arXiv preprint arXiv:1706.03825 , 2017 . D. Smilkov, N. Thorat, B. Kim, F. Vi\u00b4egas, and M.Wattenberg. Smoothgrad: removing noise by adding noise. arXiv preprint arXiv:1706.03825, 2017."},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.5555\/3305890.3306024"},{"key":"e_1_2_1_41_1","doi-asserted-by":"publisher","DOI":"10.5555\/3157382.3157633"},{"key":"e_1_2_1_42_1","first-page":"841","article-title":"Counterfactual explanations without opening the black box: Automated decisions and the gdpr","volume":"31","author":"Wachter S.","year":"2017","unstructured":"S. Wachter , B. Mittelstadt , and C. Russell . Counterfactual explanations without opening the black box: Automated decisions and the gdpr . Harv. JL & Tech. , 31 : 841 , 2017 . S. Wachter, B. Mittelstadt, and C. Russell. Counterfactual explanations without opening the black box: Automated decisions and the gdpr. Harv. JL & Tech., 31:841, 2017.","journal-title":"Harv. JL & Tech."},{"key":"e_1_2_1_43_1","volume-title":"d. Garcez. Measurable counterfactual local explanations for any classifier. arXiv preprint arXiv:1908.03020","author":"White A.","year":"2019","unstructured":"A. White and A. d. Garcez. Measurable counterfactual local explanations for any classifier. arXiv preprint arXiv:1908.03020 , 2019 . A. White and A. d. Garcez. Measurable counterfactual local explanations for any classifier. arXiv preprint arXiv:1908.03020, 2019."}],"container-title":["ACM SIGKDD Explorations Newsletter"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3468507.3468517","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3468507.3468517","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T20:17:18Z","timestamp":1750191438000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3468507.3468517"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,5,26]]},"references-count":43,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2021,5,26]]}},"alternative-id":["10.1145\/3468507.3468517"],"URL":"https:\/\/doi.org\/10.1145\/3468507.3468517","relation":{},"ISSN":["1931-0145","1931-0153"],"issn-type":[{"value":"1931-0145","type":"print"},{"value":"1931-0153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,5,26]]},"assertion":[{"value":"2021-05-29","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}