Reproduction Bundle
Paper: Incentivized Exploration with Stochastic Covariates: A Two-Stage Mechanism Design for Recommender System
OpenReview: https://openreview.net/forum?id=LTHHiPNbrs
arXiv: https://arxiv.org/abs/2406.04374
HF Job smoke check: https://huggingface.co/jobs/Srishti280992/6a5d7cc4bee6ee1cf4ed1c32
Contents
rcb/: RCB implementation from the paper pseudocode.scripts/: theorem checks, mechanism checks, synthetic runs, warfarin runs, and figure generation.data/: IWPC/PharmGKB warfarin data used by the local pipeline.outputs/: JSON and logs from the reproduced runs.figs/: Plotly HTML figures and CSV backing data.poster.html,poster_embed.html,poster_preview.png,GATE_REPORT.json: Posterly poster and gate report.
Rerun
pip install numpy pandas scikit-learn plotly xlrd
python scripts/theory_checks.py
python scripts/claim3_structure.py
python scripts/regret_rate_fit.py --seeds 5 --workers 10 --out outputs/claim1_rate.json
python scripts/run_synthetic.py --only s3 --out outputs/syn_s3.json
python scripts/run_synthetic.py --only feas --out outputs/feas.json
python scripts/run_warfarin.py --C_N 1.0 --N 5 --reward binary --EF theory --phi0 1.0 --perms 10 --out outputs/warfarin.json
python scripts/warfarin_ablation.py --perms 5 --out outputs/warfarin_ablation.json
python scripts/figures.py rate rate_Kd warfarin warfarin_ablation warfarin_N feasibility Neps tradeoff gain
The scripts chdir to the bundle root, so run them from this directory or from any parent path.