MDP/RL

Question 1

max expected utility MEU

Answer 1

max expected utility out of all actions

Question 2

what is reinforcement learning

Answer 2

MDP where P and R are not given

Question 3

what is exploration

Answer 3

• allows an agent to improve its current knowledge about each action, hopefully leading to long-term benefit.
• Improving the accuracy of the estimated action-values, enables an agent to make more informed decisions in the future.

Question 4

what is exploitation

Answer 4

• chooses the greedy action to get the most reward by exploiting the agent’s current action-value estimates.
• may not actually get the most reward and lead to sub-optimal behaviour.

Question 5

what is epsilon

Answer 5

• probability of choosing to explore
• low value: exploit action with highest expected rewards
• high value: explore new actions even with low rewards

Question 6

upper confidence-bound action selection

Answer 6

• optimism in the face of uncertainty
• a*t