release/v1/tables/selections.csv

source_batch,source_line_number,record_sha256,schema_version,experiment_id,spec_version,paper_version,openreview_id,batch_id,record_type,recorded_at,status,claim_id,panel_id,task_id,attempt_id,method,environment,seed,selection_rule,candidate_count,selected_seed,selected_summary,determinization_extension,exact_original_replay,fixed_grid_summary,reason,selection_and_reporting_grid_reused,selection_summary
hf-claim23-full-v4-20260715T1844Z,2632,4db6003836bd7cf7c331700ca01129ea20c7076bb94d7022b71a51e06105c614,1.0.0,icml2026_chebyshev_v1,1.0.0,2605.22305v4,aNWIVNjocB,hf-claim23-full-v4-20260715T1844Z,selection,2026-07-15T18:53:02.252102+00:00,success,claim3,,select_best_ch3_ars_on_fixed_start_grid,,ch3_ars,,,maximum mean return across the fixed Mountain Car start grid,20,1107,"{""count"":100,""max"":99.10945129394531,""mean"":98.95029006958008,""min"":98.74156951904297,""std"":0.09871378505667953}",,,,,true,
hf-claim4-full-v4-20260715T1845Z,2202,57db5790b09c33a0a404f287ed00dd3119b294a766accd16d566a72ab9357ad3,1.0.0,icml2026_chebyshev_v1,1.0.0,2605.22305v4,aNWIVNjocB,hf-claim4-full-v4-20260715T1845Z,selection,2026-07-15T19:19:45.531570+00:00,success,claim4,,select_best_ch3_ppo_on_fixed_start_grid,,ch3_ppo,,,maximum mean return across the fixed Mountain Car start grid,20,738,"{""count"":100,""max"":98.74390411376953,""mean"":84.58356767326593,""min"":-1.6164672374725342,""std"":34.489042480072385}",,,,,true,
hf-claim4-full-v4-20260715T1845Z,5363,eb107bbc97a42de583ca3b9fb49ecb80d07db6acc59ba43e2d898330e4b07eb2,1.0.0,icml2026_chebyshev_v1,1.0.0,2605.22305v4,aNWIVNjocB,hf-claim4-full-v4-20260715T1845Z,selection,2026-07-15T23:43:22.475191+00:00,success,claim4,,select_best_seeded_adamw_reinforce_candidate,,ch3_reinforce_adamw_seeded,,,maximum mean return over 50 common-random-number episodes,20,246,,"seeded Python, NumPy, PyTorch, and Gymnasium; common evaluation seeds",false,"{""count"":100,""max"":98.45274102933101,""mean"":96.0097169109171,""min"":-2.7467713727204854,""std"":14.006514660064246}",The paper released neither random seeds nor the selected coefficients; upstream asynchronous completion also made the stored policy index unstable.,,"{""count"":50,""max"":98.45252654356517,""mean"":98.00690990343256,""min"":97.3278696654497,""std"":0.26765168984118043}"
hf-claim5-pendulum-full-v4-20260715T1846Z,81092,db3261d4089a3d3df2846a5a0999880c80484e9ca0096891641caa59d5bfe4b3,1.0.0,icml2026_chebyshev_v1,1.0.0,2605.22305v4,aNWIVNjocB,hf-claim5-pendulum-full-v4-20260715T1846Z,selection,2026-07-16T03:00:17.188201+00:00,success,claim5,,select_best_ch6_ars_pendulum_grid,,ch6_ars_pendulum,,,maximum mean return across the 2D deterministic evaluation grid,30,3075,"{""count"":2500,""max"":-0.6707585453987122,""mean"":-150.49824872477055,""min"":-418.3816833496094,""std"":87.63166150095708}",,,,,true,