New answers tagged policies
1
vote
What happens when the probability of either one of the policies is 0 in Importance Sampling?
Very simply, one of the requirements of off-policy RL to converge, is that the behavioral policy $b$ has at least the same support of the target policy $\pi$, thus:
$$
\forall s \in S \forall a \in A \...
2
votes
What happens when the probability of either one of the policies is 0 in Importance Sampling?
Background: Importance sampling is used in many off-policy RL algorithms when the data is generated with one policy, yet it is being used to update another policy. The policy generating the data is ...
Top 50 recent answers are included
Related Tags
policies × 76reinforcement-learning × 67
markov-decision-process × 14
value-functions × 14
off-policy-methods × 7
comparison × 6
q-learning × 6
policy-gradients × 5
policy-iteration × 5
deterministic-policy × 5
deep-rl × 4
terminology × 4
dqn × 4
definitions × 4
proofs × 4
sutton-barto × 4
bellman-equations × 4
value-iteration × 4
optimal-policy × 4
optimality × 4
rewards × 3
environment × 3
importance-sampling × 3
stochastic-policy × 3
neural-networks × 2