Regional Tree Regularization for Interpretability in Deep Neural Networks

Wu, Mike and Parbhoo, Sonali and Hughes, Michael C. and Kindle, Ryan and Celi, Leo A. and Zazzi, Maurizio and Roth, Volker and Doshi-Velez, Finale. (2020) Regional Tree Regularization for Interpretability in Deep Neural Networks. In: The Thirty-Fourth AAAI Conference on Artificial Intelligence, AAAI 2020, 34. pp. 6413-6421.

Full text not available from this repository.

Official URL: https://edoc.unibas.ch/81761/

Downloads: Statistics Overview


The lack of interpretability remains a barrier to adopting deep neural networks across many safety-critical domains. Tree regularization was recently proposed to encourage a deep neural network's decisions to resemble those of a globally compact, axis-aligned decision tree. However, it is often unreasonable to expect a single tree to predict well across all possible inputs. In practice, doing so could lead to neither interpretable nor performant optima. To address this issue, we propose regional tree regularization – a method that encourages a deep model to be well-approximated by several separate decision trees specific to predefined regions of the input space. Across many datasets, including two healthcare applications, we show our approach delivers simpler explanations than other regularization schemes without compromising accuracy. Specifically, our regional regularizer finds many more “desirable” optima compared to global analogues.
Faculties and Departments:05 Faculty of Science > Departement Mathematik und Informatik > Informatik > Biomedical Data Analysis (Roth)
UniBasel Contributors:Roth, Volker and Parbhoo, Sonali
Item Type:Conference or Workshop Item, refereed
Conference or workshop item Subtype:Conference Paper
Publisher:AAAI Press
Series Name:Proceedings of the ... AAAI Conference on Artificial Intelligence
Issue Number:04
Note:Publication type according to Uni Basel Research Database: Conference paper
Identification Number:
Last Modified:10 Feb 2021 16:11
Deposited On:10 Feb 2021 16:11

Repository Staff Only: item control page