New answers tagged markov-decision-process
1
vote
Accepted
UCB, Thompson sampling etc seems myopic/greedy for bandits?
The (binary) multi-armed bandit actually is a MDP with one state and $K$ actions.
So your suggestion boils down to meta-learning: Find the parameters of one MDP that can solve another. Let's go with ...
1
vote
Markov Decision Processes Model
you have in front of you 10 slot machines
you can play with any of them, and each of them have a specific winrate (reward function)
the only state of this MDP is the initial state, the one where you ...
Top 50 recent answers are included
Related Tags
markov-decision-process × 192reinforcement-learning × 163
pomdp × 22
q-learning × 18
rewards × 14
policies × 14
reward-functions × 14
value-iteration × 12
value-functions × 11
markov-property × 10
comparison × 9
state-spaces × 9
discount-factor × 9
deep-rl × 8
definitions × 8
multi-armed-bandits × 8
transition-model × 8
machine-learning × 6
reference-request × 6
policy-gradients × 6
proofs × 6
environment × 6
sutton-barto × 6
policy-iteration × 6
dynamic-programming × 6