User Tools

Site Tools


public:t-720-atai:atai-19:engineering_assignment_2

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

public:t-720-atai:atai-19:engineering_assignment_2 [2019/10/03 09:52]
thorisson [Control Assignment 1]
public:t-720-atai:atai-19:engineering_assignment_2 [2019/10/03 09:59] (current)
thorisson [DESCRIPTION]
Line 14: Line 14:
 ====DESCRIPTION==== ====DESCRIPTION====
  
-Referring to [[public:t-720-atai:atai-19:control_assignment_1|Control Assignment 1]], implement a reinforcement learner that can do what you learned in part (a) of that assignment, i.e. learn to keep the thrust in a 1000-point range for 5 seconds. This will be done in two parts:+Referring to [[public:t-720-atai:atai-19:control_assignment_1|Control Assignment 1]], implement a reinforcement learner that can do what you learned in PART-1 of that assignment (piloting the alien space ship off the planet), i.e. learn to keep the thrust in a 1000-point range for 5 seconds. This will be done in two parts:
  
 ==Part 1== ==Part 1==
-First, write a detailed report (max 4 pages) detailing everything about how you would go about implementing such a reinforcement learner for that task. The description must be detailed enough that it could be given to your fellow student to implement, and they would not have to second-guess any key detail of the design. //Make sure to note any challenges or limitations that, compared with human performance on the task, are implied by or inherent in your design.// You may use any reinforcement learning mechanism of your choice (Q-learning, SARSA, etc.). +First, write a detailed specification/report (max 4 pages) detailing everything about how you would go about implementing such a reinforcement learner for that task. The description must be detailed enough that it could be given to your fellow student to implement, and they would not have to second-guess any key detail of the design. //Make sure to note any challenges or limitations that, compared with human performance on the task, are implied by or inherent in your design.// You may use any reinforcement learning mechanism of your choice (Q-learning, SARSA, etc.). 
  
 ==Part 2== ==Part 2==
/var/www/ailab/WWW/wiki/data/pages/public/t-720-atai/atai-19/engineering_assignment_2.txt ยท Last modified: 2019/10/03 09:59 by thorisson