CartPole reward function inconsistent with Barto et al.

[moshelooks]

In both the paper and original source code the reward function for CartPole is r = -1 when the pole falls over and r = 0 otherwise. In the gym implementation the reward is the same (r = 1), whether the pole falls over or not.

[olegklimov]

I think it's not even worth version change. Task is so, so easy. Everything works, any reward works.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.