[RL-baseline] Model v4, experiment #2

[ziritrion]

The policy network for model v4 for REINFORCE with Baseline is essentially the same network as in v2, but the actor and critic heads have an additional fully connected layer each similar to v3. This tweak was added with the hopes of seeing the initial gains in reward that we observed with model v3 but with the elevated sustained reward value that we observed in v2.

The action sets are the same as in Model v3. For this experiment, action set #1 is chosen:
[0.0, 0.0, 0.0], # no action
[0.0, 0.8, 0.0], # throttle
[0.0, 0.0, 0.6], # break
[-0.9, 0.0, 0.0], # left
[-0.5, 0.0, 0.0], # left
[-0.2, 0.0, 0.0], # left
[0.9, 0.0, 0.0], # right
[0.5, 0.0, 0.0], # right
[0.2, 0.0, 0.0], # right

The Running Reward reached its maximum value of 413 around the 4k episode mark but quickly collapsed afterwards. Entropy stayed consistently near zero at the 5k episode mark, although RR became positive again at the 7k episode mark, fluctuating between 150 and 250 up until the 17k episode mark, where the RR finally collapsed and never recovered.

Results are below:

Sample video below:
https://user-images.githubusercontent.com/1465235/113120748-328b7900-9212-11eb-8cac-2e79f1d6b90e.mp4