Description
Step 1: Problems Q1.. Q4 from Berkeley AI reinforcement learning
project (10 pts)
Solve problems Q1 to Q4 from the Berkeley AI reinforcement learning project:
http://ai.berkeley.edu/reinforcement.html
Submit:
• the changed files: analysis.py, qlearningAgents.py, valueIterationAgents.py
• the homework journal for this step.
Step 2: Visualize the evolution of the V-value (4 pts)
Re-run problem Q1 from Step 1. Plot the evolution of the V value in the upper-left corner. You can try to
do this in Excel or other spreadsheets. A better solution is to learn and use pyplot.
Submit:
• the homework journal for this step, with the last screenshot showing the plot
Step 3: Visualize the evolution of the Q-value (4 pts)
Re-run problem Q4 from Step 1. Plot the evolution of the four state action (Q) values for the second
state from left in the top row. Put all values in a single plot. Again, Excel is acceptable, but pyplot is
better.
Submit:
• the homework journal for this step, with the last screenshot showing the plot
