论文信息 - Combining Reinforcement Learning and Causal Models for Robotics Applications

Combining Reinforcement Learning and Causal Models for Robotics Applications

The relation between Reinforcement learning (RL) and Causal Modeling(CM) is an underexplored area with untapped potential for any learning task. In this extended abstract of our Ph.D. research proposal, we present a way to combine both areas to improve their respective learning processes, especially in the context of our application area (service robotics). The preliminary results obtained so far are a good starting point for thinking about the success of our research project.

Arquímides Méndez-Molina | Arquímides Méndez-Molina

[1] Silvio Savarese,et al. Causal Induction from Visual Observations for Goal Directed Tasks , 2019, ArXiv.

[2] Mathias Frisch,et al. Causation and intervention , 2014 .

[3] Tim Miller,et al. Explainable Reinforcement Learning Through a Causal Lens , 2019, AAAI.

[4] Thomas G. Dietterich. Adaptive computation and machine learning , 1998 .

[5] E. Morales,et al. Causal Based Q-Learning , 2020, Res. Comput. Sci..

[6] S. Gershman. Reinforcement Learning and Causal Models , 2017 .

[7] Zeb Kurth-Nelson,et al. Causal Reasoning from Meta-reinforcement Learning , 2019, ArXiv.

[8] Michael R. Waldmann,et al. The Oxford handbook of causal reasoning , 2017 .

[9] Hugo Jair Escalante,et al. Playing against Nature: causal discovery for decision making under uncertainty , 2018, ArXiv.

[10] Jiming Liu,et al. Incorporating causal factors into reinforcement learning for dynamic treatment regimes in HIV , 2019, BMC Medical Informatics and Decision Making.

[11] Dileep George,et al. Schema Networks: Zero-shot Transfer with a Generative Causal Model of Intuitive Physics , 2017, ICML.