959 B
959 B
REPPO Experiment Plan
Proof of Concept Success
✅ Working Implementation: https://wandb.ai/dominik_roth/reppo_dev_test?nw=nwuserdominik_roth
Experiments To Run
1. Reproduce Paper Results
✅ Brax Suite: 5 tasks - COMPLETED
- ant, cheetah, humanoid, walker, hopper
- Results: https://wandb.ai/dominik_roth/reppo_brax_production
DMC Suite (mujoco_playground): 23 tasks
- AcrobotSwingup, CartpoleBalance, CheetahRun, FingerSpin, HumanoidRun, WalkerRun, etc.
ManiSkill Suite: 8 tasks (need wrapper first)
- PickSingleYCB-v1, PegInsertionSide-v1, UnitreeG1TransportBox-v1, etc.
Settings: 50M steps, 1024 envs, 5 seeds each, paper hyperparameters
Scripts Needed
submit_experiments.py
Uses existing working SLURM script:
python submit_experiments.py --experiment brax --seeds 5
python submit_experiments.py --experiment mjx --seeds 5
python submit_experiments.py --experiment maniskill --seeds 5