I am working on a problem that can be casted as a contextual bandit problem with continuous action space. I would like to tackle it by using something like the contextual zooming algorithm from the paper Contextual Bandits with Similarity Information.

Is there any open source implementation (possibly in python) of this or other contextual bandit algorithm?

I am not an expert of MAB so I might be wrong, but it seems to me that the field lacks the equivalent of SB3 or rllib for Reinforcement Learning. I find it quite puzzling given that MAB problems are a very well studied special case of RL and they are mathematically better understood.

Any help would be very appreciated! Thanks a lot!