Causal Reinforcement Learning (Causal RL), which is a promising virgin field and will, without doubt, become an indispensable part of artificial general intelligence. The philosophy behind the integration of Causal Inference and RL is obvious but charming. That is, when looking back at the history of science, human beings always progress in a similar manner to that of Causal RL: "Humans summarise rules or experience from their interaction with nature and then exploit this to improve their adaptation in the next exploration. What CausalRL does is exactly to mimic human behaviours, learning causal relations from an agent communicating with the environment and then optimising its policy based on the learned causal structures." Therefore, Causal RL is so general that it has numerous applications in computer vision, robots, healthcare, medicine, finance, sociology, and so on. In this talk, I will discuss all these about Causal RL.
Chaochao Lu is a PhD student of Machine Learning Group at the University of Cambridge, where he is co-supervised by Prof. Zoubin Ghahramani and Prof. José Miguel Hernández-Lobato, advised by Prof. Carl Edward Rasmussen, and also jointly supervised by Prof. Bernhard Schölkopf and Dr. Michael Hirsch at the Max Planck Institute for Intelligent Systems under the Cambridge-Tübingen PhD Programme. His main research interest is causal learning, in particular causal reinforcement learning.
12 июл 2024
