In the recent years causal inference has emerged as an important approach for addressing different issues within RL. Providing agents the ability to leverage causal knowledge was identified as a key ingredient in developing human-centered explanation methods. Namely, when using AI systems, humans tend to be interested in answering questions such as “What caused the model to make this decision?” or “In what way should I change my input in order to achieve a desired output?”. These questions are inherently causal and cannot be answered by the traditional interpretable methods. This project will investigate current and develop novel approaches to exploring causality within RL, with the goal to provide explainability of RL-based models. Suitable for MSc-level students and prior experience of RL is essentiala.
School of Computer Science and Statistics