Open Butanium opened 2 years ago
In this post : https://lilianweng.github.io/posts/2018-01-23-multi-armed-bandit/
You define $r_t(a)$ but don't use it, and use $Q$ instead of $\hat{Q}_t$ in the UCB1 formula which looks like a typo
In this post : https://lilianweng.github.io/posts/2018-01-23-multi-armed-bandit/
You define $r_t(a)$ but don't use it, and use $Q$ instead of $\hat{Q}_t$ in the UCB1 formula which looks like a typo