Towards A Rigorous Science of Interpretable Machine Learning
Citations
2,827 citations
2,805 citations
2,258 citations
Cites background from "Towards A Rigorous Science of Inter..."
...Available: https://christophm.github.io/interpretable-ml-book/ [106] O. Bastani, C. Kim, and H. Bastani....
[...]
...Available: https://sites.google.com/view/whi2018/ [6] A. G. Wilson, B. Kim, and W. Herlands....
[...]
...Doshi-Velez and Kim established a baseline of evaluation approaches and proposed three major types of interpretability evaluation: (i) application-grounded: put the explanation into the application and let the end user (typically a domain expert) test it....
[...]
...Most research works on the ML interpretability agreed and contribute towards more rigorous notion of interpretability [62]....
[...]
...[134] B. Kim, C. Rudin, and J. A. Shah, ‘‘The Bayesian case model: A generative approach for case-based reasoning and prototype classification,’’ in Proc....
[...]
1,602 citations
1,477 citations
References
14,377 citations
"Towards A Rigorous Science of Inter..." refers background in this paper
...…to predictive policing systems, machine learning (ML) systems are increasingly ubiquitous; they outperform humans on specific tasks [Mnih et al., 2013, Silver et al., 2016, Hamill, 2017] and often guide processes of human understanding and decisions [Carton et al., 2016, Doshi-Velez et al., 2014]....
[...]
12,940 citations
"Towards A Rigorous Science of Inter..." refers background in this paper
...Just as there are now large open repositories for problems in classification, regression, and reinforcement learning [Blake and Merz, 1998, Brockman et al., 2016, Vanschoren et al., 2014], we advocate for the creation of repositories that contain problems corresponding to real-world tasks in which…...
[...]
11,104 citations
8,757 citations
"Towards A Rigorous Science of Inter..." refers background in this paper
...…email-filters to predictive policing systems, machine learning (ML) systems are increasingly ubiquitous; they outperform humans on specific tasks [Mnih et al., 2013, Silver et al., 2016, Hamill, 2017] and often guide processes of human understanding and decisions [Carton et al., 2016,…...
[...]
2,690 citations
"Towards A Rigorous Science of Inter..." refers background in this paper
...…criteria such as safety [Otte, 2013, Amodei et al., 2016, Varshney and Alemzadeh, 2016], nondiscrimination [Bostrom and Yudkowsky, 2014, Ruggieri et al., 2010, Hardt et al., 2016], avoiding technical debt [Sculley et al., 2015], or providing the right to explanation [Goodman and Flaxman, 2016]....
[...]
...• Multi-objective trade-offs: Two well-defined desiderata in ML systems may compete with each other, such as privacy and prediction quality [Hardt et al., 2016] or privacy and nondiscrimination [Strahilevitz, 2008]....
[...]