It is shown that the randomized stationary policy is an optimal policy in m (c) if and only if it is convex combination of some deterministic stationary optimal policies.

特别是证明了：一随机平稳策略，它在(c)上是最优的充要条件是它可表为若干个决定性平稳最优策略的凸组合。

youdao

It is shown that the randomized stationary policy is an optimal policy in m (c) if and only if it is convex combination of some deterministic stationary optimal policies.

特别是证明了：一随机平稳策略，它在(c)上是最优的充要条件是它可表为若干个决定性平稳最优策略的凸组合。

youdao