# Conclusion --- ## Reproduction bundle Workspace: `metalearning-quantum-control/` (this directory) containing official code + `repro_scripts/`, `outputs/`, checkpoints (`temp_maml_claim2*.pt`), LQR/PL figures, and Trackio logbook. ### How to rerun ```bash cd metalearning-quantum-control uv sync uv run python experiments/figs_appendix_classical/fig_a1_a2_lqr_simplified.py # Claim 6 uv run python experiments/fig_2_lemma_validation/lemma_validation.py # Claim 4 uv run python repro_scripts/eval_claim2_from_ckpt.py # Claim 2 uv run python repro_scripts/eval_claim5_from_ckpt.py # Claim 5 # Claim 3 (GPU): hf jobs uv run --flavor a10g-small --timeout 2h --secrets HF_TOKEN \ --env CODE_ROOT=/mnt/code --env OUT_ROOT=/mnt/outputs \ -v hf://buckets/Sor0ush/icml-repro-metaqctrl-artifacts:/mnt \ repro_scripts/hf_job_claim3.py ``` ### Hub artifacts - Jobs: https://huggingface.co/jobs/Sor0ush/6a5badf4d216bd6f3a1fedda (CZ GPU), https://huggingface.co/jobs/Sor0ush/6a5b49ccd216bd6f3a1fded1 (prior CZ) - Bucket: https://huggingface.co/buckets/Sor0ush/icml-repro-metaqctrl-artifacts - Paper: https://huggingface.co/papers/2601.18973 · https://openreview.net/forum?id=Gu9pw3D6vT --- **🎯 Trackio dashboard** `repro-when-does-adaptation-win-scaling-laws-for-meta-learning-in-quantum-control` trackio-local-dashboard://repro-when-does-adaptation-win-scaling-laws-for-meta-learning-in-quantum-control --- **📦 Artifact** `repro-when-does-adaptation-win-scaling-laws-for-meta-learning-in-quantum-control/repro-bundle:v0` · dataset https://huggingface.co/buckets/Sor0ush/repro-when-does-adaptation-win-scaling-laws-for-meta-learning-in-quantum-control-artifacts#repro-when-does-adaptation-win-scaling-laws-for-meta-learning-in-quantum-control/repro-bundle:v0