I made a Monte Carlo tree search (MCTS) algorithm for the travelling salesman problem inspired by this paper which uses UCB1.
When I was digging to see where does the UCB1 formula comes from, I read this article which then introduces the Thompson sampling.
I was wondering if in a MCTS for a problem like the TSP, the Thompson sampling could be used (Most of the usage I saw of Thompson sampling with MCTS were with in Bernoulli configuration for exemple this one in which UCB1 performs better) and if it would be more efficient than the UCB1.
