mathlab , Computer Engineering

Assignment Help:
Windy Grid World

This assignment is to use Reinforcement Learning to solve the following "Windy Grid World" problem illustrated in the above picture. Each cell in the image is a state. There are four actions: move up, down, left, and right. This is a deterministic domain -- each action deterministically moves the agent one cell in the direction indicated. If the agent is on the boundary of the world and executes an action that would move it "off" of the world, it remains on the grid in the same cell from which it executed the action.
Notice that there are arrows drawn in some states in the diagram. These are the "windy" states. In these states, the agent experiences an extra "push" upward. For example, if the agent is in a windy state and executes an action to the left or right, the result of the action is to move left or right (respectively) but also to move one cell upward. As a result, the agent moves diagonally upward to the left or right.
This is an episodic task where each episode lasts no more than 30 time steps. At the beginning of each episode, the agent is placed in the "Start" state. Reward in this domain is zero everywhere except when the agent is in the goal state (labeled "goal" in the diagram). The agent receives a reward of positive ten when it executes any action {\it from} the goal state. The episode ends after 30 time steps or when the agent takes any action after having landed in the goal state.
You should solve the problem using Q-learning. Use e-greedy exploration with epsilon=0.1 (the agent takes a random action 10 percent of the time in order to explore.) Use a learning rate of 0.1 and a discount rate of 0.9.
The programming should be done in MATLAB. Students may get access to MATLAB here. Alternatively, students may code in Python (using Numpy). If the student would rather code in a different language, please see Dr Platt or the TA.
Students should submit their homework via email to the TA ([email protected]) in the form of a ZIP file that includes the following:
1. A PDF of a plot of gridworld that illustrates the policy and a path found by Q-learning after it has approximately converged. The policy plot should identify the action taken by the policy in each state. The path should begin in the start state and follow the policy to the goal state.
2. A PDF of a plot of reward per episode. It should look like the diagram in Figure 6.13 in SB.
3. A text file showing output from a sample run of your code.
4. A directory containing all source code for your project.
5. A short readme file enumerating the important files in your submission.
Updates
You can initialize the Q function randomly or you can initialize it to a uniform value of 10. That is, you can initialize Q such that each value in the table is equal to 10.
There have been questions about how to know when the algorithm has converged. The algorithm has converged when the value function has stopped changing signficantly and the policy has stopped changing completely. Since we are using q-learning, the algorithm should converge to a single optimal policy.
Please also submit a short readme file with your homework that enumerates the important files in your submission.


Related Discussions:- mathlab

Programmed input - output technique for computers, Q. Programmed input - ou...

Q. Programmed input - output technique for computers? Programmed input/output is a useful I/O technique for computers where hardware costs need to be minimised. Input or output

What is a diagonalization language ld, What is a Diagonalization language L...

What is a Diagonalization language LD?  The  diagonalization  language  having  of  all  strings  w  such  that  the  TM  whose  code is w does not accept when w is given as in

Dialog-task updates, Dialog-task updates are Synchronous updates.

Dialog-task updates are Synchronous updates.

How do you get workflow automation in e-business environment, How do you ge...

How do you get workflow automation into e-business environment? In order to run smoothly, organizations frequently standardize processes across the organization and support use

What are the methods for handling deadlocks, What are the methods for handl...

What are the methods for handling deadlocks?  The deadlock problem can be dealt with in one of the three ways:  a. Use a protocol to prevent or avoid deadlocks, make sure th

Demux., Implement the full substractor using 8:1demux

Implement the full substractor using 8:1demux

Give the solutions for critical section problem, Give the solutions for cri...

Give the solutions for critical section problem? A solution to the critical section problem must specify the subsequent requirements. (a) Mutual exclusion (b) Progress (c)

Explain status and control registers, Q. Explain Status and Control Registe...

Q. Explain Status and Control Registers? For control of numerous operations several registers are used. These registers can't be used in data manipulation though content of som

Explain how server form post-back works, Briefly explain how server form po...

Briefly explain how server form post-back works?  Post Back: The process in which a Web page sends data back to the similar page on the server. View State: View State is the m

Explian two limitations of dead-box analysis, (a) Explian two limitations o...

(a) Explian two limitations of dead-box analysis. (b) Describe why memory analysis is difficult. (c) With reference to the "Shadow Walker" rootkit, explain what is meant by

Write Your Message!

Captcha
Free Assignment Quote

Assured A++ Grade

Get guaranteed satisfaction & time on delivery in every assignment order you paid with us! We ensure premium quality solution document along with free turntin report!

All rights reserved! Copyrights ©2019-2020 ExpertsMind IT Educational Pvt Ltd