MuJoCo Hopper Environment

Overview

Make a two-dimensional one-legged robot hop forward as fast as possible.

Performances of RL Agents

We list various reinforcement learning algorithms that were tested in this environment. These results are from RL Database. If this page was helpful, please consider giving a star!

Star

Result Algorithm Source
3915.9 ACKTR Scalable trust-region method for deep reinforcement learning using Kronecker-factored approximation
3915.3 A2C Scalable trust-region method for deep reinforcement learning using Kronecker-factored approximation
3804.9 Trust-PCL Trust-PCL: An Off-Policy Trust Region Method for Continuous Control
3765.78 TRPO+GAE Trust-PCL: An Off-Policy Trust Region Method for Continuous Control
3755.0 TRPO Scalable trust-region method for deep reinforcement learning using Kronecker-factored approximation
3564.07 TD3 Addressing Function Approximation Error in Actor-Critic Methods
3516.7 TRPO Trust-PCL: An Off-Policy Trust Region Method for Continuous Control
2996.66 SAC Addressing Function Approximation Error in Actor-Critic Methods
2471.3 TRPO Addressing Function Approximation Error in Actor-Critic Methods
2428.39 ACKTR Addressing Function Approximation Error in Actor-Critic Methods
2316.16 PPO OpenAI Baselines ea68f3b
2164.7 PPO Addressing Function Approximation Error in Actor-Critic Methods
2020.46 DDPG Addressing Function Approximation Error in Actor-Critic Methods
1912.9 TRPO (MPI) OpenAI Baselines ea68f3b
1860.02 Our DDPG Addressing Function Approximation Error in Actor-Critic Methods