Hamid Eghbal-Zadeh, Florian Henkel, Gerhard Widmer,
"Context-Adaptive Reinforcement Learning using Unsupervised Learning of Context Variables"
, in PMLR: NeurIPS 2020 Workshop on Pre-registration in Machine Learning, Serie PMLR - Proceedings of Machine Learning Research, Vol. 148, Seite(n) 236-254, 12-2021
Original Titel:
Context-Adaptive Reinforcement Learning using Unsupervised Learning of Context Variables
Sprache des Titels:
Englisch
Original Buchtitel:
NeurIPS 2020 Workshop on Pre-registration in Machine Learning
Original Kurzfassung:
{In Reinforcement Learning (RL), changes in the
context often cause a distributional change in the observations of the
environment, requiring the agent to adapt to this change. For example,
when a new user interacts with a system, the system has to adapt to the
needs of the user, which might differ based on the user?s
characteristics that are often not observable. In this Contextual
Reinforcement Learning (CRL) setting, the agent has to not only
recognise and adapt to a context, but also remember previous ones.
However, often in CRL the context is unknown, hence a supervised
approach for learning to predict the context is not feasible. In this
paper, we introduce Context-Adaptive Reinforcement Learning Agent
(CARLA), that is capable of learning context variables in an
unsupervised manner, and can adapt the policy to the current context. We
provide a hypothesis based on the generative process that explains how
the context variable relates to the states and observations of an
environment. Further, we propose an experimental protocol to test and
validate our hypothesis; and compare the performance of the proposed
approach with other methods in a CRL environment. Finally, we provide
empirical results in support of our hypothesis, demonstrating the
effectiveness of CARLA in tackling CRL.
Forschungs-