Optimization Methods for Interpretable Differentiable Decision Trees Applied to Reinforcement Learning

By Andrew Silva, Matthew Craig Gombolay, Taylor W. Killian, I. Jimenez, Sung-Hyun Son. 2020

115 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Yashesh D. Dhebar
2. Subramanya P. Nageshrao
3. Ling Zhu
4. Dimitar Filev
5. Kalyanmoy Deb
6. Vinícius G. Costa
7. Carlos Eduardo Pedreira
8. Yaochu Jin

Citations in the corpus

• Toward Interpretable-AI Policies Using Evolutionary Nonlinear Decision Trees for Discrete-Action Systems (2020, methodology): “In [29], a gradient-based approach is developed to train the DT of pre-fixed topology involving linear split rules.”
• Recent advances in decision trees: an updated survey (2022, methodology): “An intuitive solution is to simply revert to the traditional structure: Silva et al. (2020) proposed inducing a multi-variate soft tree using gradient-based techniques and then transforming it into a univariate crisp tree by selecting the attributes with higher weights.”