How should I decay $\epsilon$ in Q-learning?

Question

How should I decay the $\epsilon$ in Q-learning?

Currently, I am decaying epsilon as follows. I initialize $\epsilon$ to be 1, then, after every episode, I multiply it by some $C$ (let it be $0.999$), when it reaches $0.01$. After that, I keep $\epsilon$ to be $0.01$ all the time. I think this has a terrible consequence.

So, I need a $\epsilon$ decay algorithm. I haven't found script or formula about it, so can you tell me?

Answer 1

The way you have described tends to be the common approach. There are of course other ways that you could do this e.g. using an exponential decay, or to only decay after a 'successful' episode, albeit in the latter case I imagine you would want to start with a smaller $\epsilon$ value and then decay by a larger amount.