Scribd
Upload
2 views1 page

Q Learning Notes

Q-learning is a model-free reinforcement learning algorithm aimed at maximizing cumulative rewards by learning the optimal action-value function. The algorithm involves initializing a Q-table, observing the current state, choosing actions using an ε-greedy strategy, and updating Q-values based on received rewards until a goal is reached. The process is illustrated in a flowchart that outlines the steps from initialization to goal completion.

Uploaded by

satpal83421
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
2 views1 page

Q Learning Notes

Q-learning is a model-free reinforcement learning algorithm aimed at maximizing cumulative rewards by learning the optimal action-value function. The algorithm involves initializing a Q-table, observing the current state, choosing actions using an ε-greedy strategy, and updating Q-values based on received rewards until a goal is reached. The process is illustrated in a flowchart that outlines the steps from initialization to goal completion.

Uploaded by

satpal83421
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Q-Learning in Artificial Intelligence

Definition: Q-learning is a model-free reinforcement learning algorithm that learns the optimal
action-value function to maximize cumulative reward.

Q-Learning Formula:
Q(s,a) = Q(s,a) + α [r + γ max Q(s',a') - Q(s,a)]

Algorithm Steps:
1. Initialize Q-table with zeros
2. Observe current state
3. Choose action using ε-greedy
4. Take action and receive reward
5. Update Q-value
6. Move to next state
7. Repeat until goal is reached

Flowchart:
Start
|
Initialize Q-table
|
Observe State (s)
|
Choose Action (ε-greedy)
|
Take Action → Reward → Next State
|
Update Q(s,a)
|
Goal reached?
| Yes → End
| No → Repeat

You might also like