The policy denoted by prescribes what action is to be taken given the state. It can be seen as a function that maps states to actions. There are two major types of policy: deterministic policies and stochastic policies.
A deterministic policy prescribes one action for a given state, that is, there is only one action, , given s. Mathematically, it means .
A stochastic policy prescribes an action distribution given a state at time step ...