Q Learning Algorithm شرح

[@roozy_313a2]

رحم الله والديك دكتور 🌺

[@MK-cu1se]

❤❤❤

[@shahd.h6430]

episode 1: state= room 4, action-room =0 Q(4,0)=100+ 0.1*0 =100 episode 2: state=3, action=4 Q(3,4)=0+0.1*100=10 episode 3: state=1, action=3 Q(1,3)=0+0.1*10=1 So the path will be from room 1 to room 3 to room 4 to room 0 which is the goal

[@youssef-ns9ny]

Youssef mohamed The quiz consists of three episodes, where the goal is to determine the values of certain Q-functions. In episode 1, the Q-function for state (4,0) is determined to be 100, which signals the completion of the episode. In episode 2, the Q-function for state (5,4) is found to be 10, with the Q-function for state (4,0) still being 100, and the episode is completed. Similarly, in episode 3, the Q-function for state (3,4) is found to be 10, with the Q-function for state (4,0) still being 100, and the episode is completed. The final path to reach the goal room 0 involves moving from room 1 to room 3 to room 4 before reaching the goal in room 0.

[@yosrmahmod5473]

Episode 1: Initial state at room 1 Q(1,3)=0+0.1*0=0 Q(3,4)=0+0.1*100=10 Episode 2: Initial state at room 4 Q(4,0)=100+0.1*0=100 Episode 3: Initial state at room 5 Q(5,4)=0+0.1*100=10 Path 5 to 4 to 0

[@jawaheralbaddawi]

what we do if start from room 0 ?

[@user-ie3hr9ez9u]

Episode 1 Q(4,0) = 100+0.1*0 =100 Episode 2 Q(3,4) = 0 + 0.1*100= 10 Episode 3 Q(1,3) = 0+0.1*10 = 1 So from room 1 to 3 to 4 to 0 and that's our goal

[@user-mu7ht8lj1f]

Abdallah Ali Ahmed In episode 1, the quiz begins with the initial state at room 1. The Q-function for state (1,3) is calculated to be 0, where the reward obtained from moving from state (1,3) to state (3,4) is 0, and the discount factor is 0.1. On the other hand, the Q-function for state (3,4) is determined to be 10, where the reward obtained from moving from state (3,4) to terminal state (4,0) is 100, and the discount factor is 0.1. In episode 2, the quiz starts with the initial state at room 4. The Q-function for state (4,0) is calculated to be 100, where the reward obtained from moving from state (4,0) to terminal state (4,0) is 0, and the discount factor is 0.1. In episode 3, the quiz begins with the initial state at room 5. The Q-function for state (5,4) is determined to be 10, where the reward obtained from moving from state (5,4) to terminal state (4,0) is 100, and the discount factor is 0.1. To reach the goal at room 0, the final path involves moving from state (5,4) to state (3,4) to state (4,0). final path from 5 to 4 to 0, Thank you,

[@nadaabdo7830]

Nada Abdelregal The answer of the quiz is: Episode 1: State:room4 , action: room 0 Q(4,0)=100 which is the goal Episode 2: State:room5 , action: room 4 Q(5,4)=0+0.1*(100 ,0 ,0)=10 Episode 3: State:room3 , action: room 4 Q(3,4)=0+0.1*(100 ,0 ,0)=10 So the path will be from room 1 to room 3 to room 4 to room 0 which is the goal

[@AhmedAli-sz9vi]

Ahmed Ali : the answer of quiz : episode 1: Q(4,0)=100 (goal) Then finish the episode episode 2: Q(5,4)=10 Q(4,0)=100 (goal) Then finish the episode episode 3: Q(3,4)=10 Q(4,0)=100 (goal) Then finish the episode and the path in the end will be from room 1 to room 3 to room 4 to the goal room 0

[@mouradfakhfakh4823]

Would you please share the python code. Thanks!

[@user-cr3ly1xy8b]

Mai Mohamed The answer of the quiz is: Episode 1: State:room4 , action: room 0 Q(4,0)=100 which is the goal Episode 2: State:room5 , action: room 4 Q(state, action)=R(state, action)+Gamma*max(Q(next state) Q(5,4)=0+0.1*(100 ,0 ,0)=10 Episode 3: State:room3 , action: room 4 Q(state, action)=R(state, action)+Gamma*max(Q(next state) Q(3,4)=0+0.1*(100 ,0 ,0)=10 So the path will be from room 1 to room 3 to room 4 to room 0 which is the goal

[@user-zu3xc8bn2k]

episode 1: state= room 4 , action=room =0 Q(4,0)=100 + 0.1*0 =100 episode 2: state=5 , action=1 Q(5,1)= 0 + 0.1*0 =0 episode 3: state=3 , action=4 Q(3,4)= 0 + 0.1*100 =10 Then finish the episode

[@manarmohamed2923]

Manar Mohamad The Answer Of the Quiz is : Episode 1 Random Stata room 3 action 4 Update Q Matrix Q(3,4)=0+0.1*100=10 4 not a goal Then choose state 4 action 0 Q(4,0)=100+0.1*0=100 0 is a goal - > Episode 1 Finished -------------------------- Episode 2 Random State room 5 action 4 Update Q Matrix Q(5,4)=0+0.1*100=10 4 not a goal Then choose state 4 action 0 Q(4,0)=100+0.1*0=100 0 is a goal - > Episode 2 Finished -------------------------- Episode 3 Random State room 4 action 0 Update Q Matrix Q(4,0)=100+0.1*0=100 0 is a goal - > Episode 3 Finished -------------------------- Start From Stare 5 by choosing Max value Optimal path is 5 to 4 Then From 4 to 0 [5-4-0]

[@MoatazMEwis]

Here is the quiz, prof. Mohamed: drive.google.com/file/d/12cYcErLIkQWl3pyGn-ya5N6hSftSJBTp/view?usp=drivesdk

[@gamaladel9308]

❤❤