Reproducing results of the paper on Mujoco domain

[carpedm20]

Working on paper branch (link).

environment	Best return for 200 steps
InvertedPendulum-v1
InvertedDoublePendulum-v1
Reacher-v1
HalfCheetah-v1	100
Swimmer-v1
Hopper-v1
Walker2d-v1
Ant-v1
Humanoid-v1
HumanoidStandup-v1

[carpedm20]

HalfCheetah-v1 is trainable with checkpoints/env_name=HalfCheetah-v1/action_fn=tanh/action_w=uniform_big/batch_size=100/clip_action=False/discount=0.99/hidden_dims=[200,200]/hidden_fn=tanh/hidden_w=uniform_big/learning_rate=0.0001/max_episodes=10000/max_steps=150/noise=ou/noise_scale=0.3/tau=0.001/update_repeat=5/use_batch_norm=False/use_seperate_networks=False/w_reg=none/w_reg_scale=0.001

[andrewliao11]

hi @carpedm20 thank for your great implementation, but I wonder if there's any other results for Mujoco benchmark

[carpedm20]

Sorry but I didn't test this on Mujoco and I don't have any plan for this project.