You can deploy this model and start earning money today!

TUM-AVS/
ikce-walker-walk-artifacts

Name: ikce-walker-walk-artifacts
Availability: InStock

reinforcement-learning

2607.05966

mit

Views

No views yet

iKCE diagnostic — walker-walk DreamerV3 artifacts

Trained checkpoints and saved rollouts that back the workshop paper Imagined Rollouts are Kinematic, Not Dynamic: A Diagnosis of Long-Horizon World-Model Failure (Robot World Model Workshop @ RSS 2026).

Trained DreamerV3 walker-walk policy in real physics: mu=1.0 walks, mu=0.1 collapses

The trained DreamerV3 policy rolled out in real physics (walker-walk), at the training friction (μ = 1.0, walks) and below the regime boundary (μ = 0.1, collapses). The world model is proprioceptive-only and does not render pixels — these clips show the environment, not the WM's imagination.

This repo is consumed by the code at github.com/TUM-AVS/iKCE; see docs/CHECKPOINTS.md there for the full provenance and the training commands. Fetch into a clone via the included scripts/fetch_artifacts.sh, or manually with hf download <this-repo> --local-dir . from the repo root.

checkpoints/
├── walker_walk/              # default checkpoint (seed 0, imag_horizon=15)
├── walker_walk_h64/          # actor-training-horizon ablation (imag_horizon=64)
├── walker_walk_seed1/        # seed-variability replicate (seed 1)
├── walker_walk_seed2/        # seed-variability replicate (seed 2)
└── walker_walk_dr/           # domain-randomization control
                              # (μ ∼ U(0.1, 1.7) per-episode at training time)

results/
├── walker_walk_wm_h64/                # WM-imagined rollouts under default actor
├── walker_walk_physics_policy_h64/    # physics rollouts under default actor
├── walker_walk_wm_h64_dr/             # WM-imagined rollouts under DR policy
└── walker_walk_physics_policy_h64_dr/ # physics rollouts under DR policy

Each checkpoints/<run>/ ships:

latest.pt — DreamerV3 weights (~219 MB), SHA-256 in docs/CHECKPOINTS.md
config.yaml — resolved config the iKCE adapter loads at inference time
metrics.jsonl — training metrics (per-eval train_return / eval_return)

Each results/<run>/<axis>/ ships:

rollout_NNN.pt — saved per-rollout state trajectories (friction: 20 rollouts per μ × 13 cells = 260; joint_noise: 20 × 10 cells = 200 per run-axis pair)
ikce_summary.csv — bootstrap-aggregated iKCE per cell

License

MIT. See the LICENSE in the code repository.

Citation

If you use these artifacts, please cite the workshop paper. A machine-readable CITATION.cff is included in this repository — Hugging Face renders it as the "Cite this repository" box above — and also lives in the code repository.

TUM-AVS/ikce-walker-walk-artifacts

iKCE diagnostic — walker-walk DreamerV3 artifacts

Contents

License

Citation

TUM-AVS/
ikce-walker-walk-artifacts