Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

MuJoCo Hopper dataset and benchmarks – Page 2 · SOTA2 Research

Research Datasets

MuJoCo Hopper

Benchmarks

Task NameDataset NameSOTA MethodSort

Task NameDataset NameSOTA ResultTrendResults

Continuous Control

MuJoCo Hopper v3 (1M steps)

3,687Average Return

5

Continuous Control

MuJoCo Hopper v3 (500K steps)

3,548Average Return

5

Policy Optimization

MuJoCo Hopper H=40

71Return

5

Continuous Control

MuJoCo Hopper (H=40)

71Normalized Return

5

Multi-objective Reinforcement Learning

MuJoCo Hopper-3

3.889Hypervolume (HV)

4

Mass Generalization

MuJoCo Hopper 1.5–2.0× mass

62.4Retention Rate

4

Dynamics Model Prediction

MuJoCo Hopper medium-replay v2 (test)

0.408RMSE

4

Dynamics Model Prediction

MuJoCo Hopper expert v2 (test)

0.322RMSE

4

Dynamics Model Prediction

MuJoCo Hopper medium v2 (train)

0.034RMSE

4

Reinforcement Learning

MuJoCo Hopper unseen (test)

1,673.58Average Return

3

Reinforcement Learning

MuJoCo Hopper-Medium (HP-M)

91.2Normalized Score

3

MuJoCo Hopper (t = T)

689.6Average Return

3

MuJoCo Hopper t = 3T/4

516.4Average Return

3

MuJoCo Hopper t = 2T/3 (shift)

397.5Average Return

3

MuJoCo Hopper t = T/2 (shift)

211.8Average Return

3

Policy Gradient

MuJoCo Hopper (final 20 iterations)

185.9Average Return

3

Continuous Control

MuJoCo Hopper v2 (test)

2,400LUTs

2

Continuous Control

MuJoCo Hopper D4RL medium

1,911.12Return

2

Page 2 of 2

Popular tasks

Continuous Control Locomotion Reinforcement Learning Offline Reinforcement Learning Dynamics Model Prediction

Follow for updates

Join SOTA2 Contact