{"type":"ScholarlyArticle","title":"A Compliance Checking Framework for DNN Models","authors":[{"name":"Sunny Verma","slug":"sunny-verma","given":"Sunny","family":"Verma","affiliation":"Advanced Analytics Institute , School of Computer Science , University of Technology Sydney , Australia; Commonwealth Scientific and Industrial Research Organisation , CSIRO , Data61 , Sydney , Australia"},{"name":"Chen Wang","slug":"chen-wang","given":"Chen","family":"Wang","affiliation":"Commonwealth Scientific and Industrial Research Organisation , CSIRO , Data61 , Sydney , Australia"},{"name":"Liming Zhu","slug":"liming-zhu","given":"Liming","family":"Zhu","affiliation":"Commonwealth Scientific and Industrial Research Organisation , CSIRO , Data61 , Sydney , Australia"},{"name":"Wei Liu","slug":"wei-liu","given":"Wei","family":"Liu","affiliation":"Advanced Analytics Institute , School of Computer Science , University of Technology Sydney , Australia"}],"abstract":"Growing awareness towards ethical use of machine learning (ML) models has created a surge for the development of fair models. Existing work in this regard assumes the presence of sensitive attributes in the data and hence can build classifiers whose decisions remain agnostic to such attributes. However, in the real world settings, the end-user of the ML model is unaware of the training data; besides, building custom models is not always feasible. Moreover, utilizing a pre-trained model with high accuracy on certain dataset can not be assumed to be fair. Unknown biases in the training data are the true culprit for unfair models (i.e., disparate performance for groups in the dataset). In this preliminary research, we propose a different lens for building fair models by enabling the user with tools to discover blind spots and biases in a pre-trained model and augment them with corrective measures.","doi":"10.24963/ijcai.2019/924","url":"https://ijcai.org/proceedings/2019/924","pdf":"https://ijcai.org/proceedings/2019/0924.pdf","markdown":"https://ijcai.org/proceedings/2019/924.md","bibtex":"https://ijcai.org/proceedings/2019/bibtex/924","volume":{"slug":"2019","title":"Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2019","url":"https://ijcai.org/proceedings/2019/","year":2019,"location":"Macao","dates":"10-16 August 2019","editors":"Sarit Kraus","isbn":"978-0-9992411-4-1","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2019"},"section":"Doctoral Consortium","track":"Doctoral Consortium","number":924,"pages":{"first":6470,"last":6471},"keywords":["Machine Learning: Interpretability","Machine Learning: Explainable Machine Learning","Humans and AI: Ethical Issues in AI"],"videos":[],"references":[{"position":1,"raw":"[Buolamwini and Gebru, 2018] Joy Buolamwini and Timnit Gebru. Gender shades: Intersectional accuracy dispari- ties in commercial gender classification. In Conference on Fairness, Accountability and Transparency, 2018.","title":"Joy Buolamwini and Timnit Gebru. Gender shades: Intersectional accuracy disparities in commercial gender classification","authors":"Gebru Buolamwini","year":"2018","venue":"Conference on Fairness, Accountability and Transparency","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Kim et al., 2018] Been Kim, Martin Wattenberg, Justin Gilmer, Carrie Cai, James Wexler, Fernanda Viegas, et al. Interpretability beyond feature attribution: Quantitative testing with concept activation vectors (tcav). In Inter- national Conference on Machine Learning, pages 2673- 2682, 2018.","title":"Interpretability beyond feature attribution: Quantitative testing with concept activation vectors (tcav)","authors":"Kim","year":"2018","venue":"International Conference on Machine Learning","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Kleinberg et al., 2019] Jon Kleinberg, Jens Ludwig, Send- hil Mullainathan, and Cass R Sunstein. Discrimination in the age of algorithms. Technical report, National Bureau of Economic Research, 2019.","title":"Discrimination in the age of algorithms","authors":"Kleinberg","year":"2019","venue":null,"doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Torralba and Efros, 2011] Antonio Torralba and Alexei A Efros. Unbiased look at dataset bias. In Computer Vision and Pattern Recognition, IEEE Conference on, 2011.","title":"Unbiased look at dataset bias","authors":"Efros Torralba, Antonio Torralba, Alexei A Efros","year":"2011","venue":"Computer Vision and Pattern Recognition","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}