Distilling Deep Reinforcement Learning Policies in Soft Decision Trees

By Youri Coppens, Kyriakos Efthymiadis, Tom Lenaerts, Ann Nowé. 2019

115 citations recorded by the source corpus.

Subfields: Reinforcement Learning, Efficiency & Compression

How researchers see this paper

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

In Reinforcement Learning

1. Nathan Dahlin
2. Pierluigi Nuzzo
3. Krishna Chaitanya Kalagarla
4. Nikhil Vijay Naik
5. Zhi Chen
6. Cynthia Rudin
7. Rahul Jain

In Efficiency & Compression

1. Nathan Dahlin
2. Pierluigi Nuzzo
3. Krishna Chaitanya Kalagarla
4. Nikhil Vijay Naik
5. Rahul Jain

Citations in the corpus

• Interpretable Machine Learning: Fundamental Principles and 10 Grand Challenges (2021, background): “, posthoc approximations), including tree-based explanations (Coppens et al., 2019; Liu et al., 2018), reward decomposition (Juozapaitis et al.”
• Designing Interpretable Approximations to Deep Reinforcement Learning with Soft Decision Trees (2020, positive): “Much recent work on distillation techniques and their application focuses on controller performance (Bastani, Pu, and Solar-Lezama 2018), (Coppens et al. 2019), (Frosst and Hinton 2017).”
• Explainable Reinforcement Learning: A Survey (2020, unknown): “et al. [54]) Post-hoc • Genetic Programming (Hein et al. [23]) • Reward Decomposition (Juozapaitis et al. [27]) • Expected Consequences (van der Waa et al. [64]) • Soft Decision Trees (Coppens et al. [8]) • Deep Q-Networks (Zahavy et al. [66]) • Autonomous Policy Explanation (Hayes and Shah [21]) • ”
• Explainable Reinforcement Learning Through a Causal Lens (2019, methodology): “Soft decision trees have been used to generate more interpretable policies (Coppens et al. 2019), and reward decomposition has been utilized to provide minimum sufficient explanations in RL (Juozapaitis et al.”