Skip to main content

Mountain Car Offline Experiments

Created on February 2|Last edited on February 4

Comparing training performance with and without human actions:

-2 Penalty, 600,000 Iterations, over 5 Seeds Each (with and without human actions)


Select runs that logged test_score
to visualize data in this line chart.
Run set
0


-1 Penalty, 600,000 Iterations, over 5 Seeds Each (with and without human actions)


Run set
0