{ "bench_version": 1, "track": "graph_landmark_forecast", "model": "transformer_tiny", "metric_direction": "lower is better", "n_seeds": 8, "baseline": { "best_cfg": { "lr": 0.003 }, "sweep": [ { "cfg": { "lr": 0.001 }, "mean": 0.09760121069848537 }, { "cfg": { "lr": 0.003 }, "mean": 0.09715960919857025 }, { "cfg": { "lr": 0.01 }, "mean": 0.09787633270025253 } ], "full": { "mean": 0.09728806372731924, "std": 0.0028138072522287676, "per_seed": [ 0.0970306470990181, 0.09655177593231201, 0.09898804873228073, 0.09606796503067017, 0.10214304178953171, 0.09961570054292679, 0.09206581115722656, 0.09584151953458786 ], "n": 8 }, "full_grid": [ { "cfg": { "lr": 0.001 }, "result": { "mean": 0.09813422616571188, "std": 0.003308826432344093, "per_seed": [ 0.09822424501180649, 0.09727376699447632, 0.0995187908411026, 0.09538803994655609, 0.10483681410551071, 0.10019651055335999, 0.09305645525455475, 0.0965791866183281 ], "n": 8 } }, { "cfg": { "lr": 0.003 }, "result": { "mean": 0.09728806372731924, "std": 0.0028138072522287676, "per_seed": [ 0.0970306470990181, 0.09655177593231201, 0.09898804873228073, 0.09606796503067017, 0.10214304178953171, 0.09961570054292679, 0.09206581115722656, 0.09584151953458786 ], "n": 8 } }, { "cfg": { "lr": 0.01 }, "result": { "mean": 0.09821864496916533, "std": 0.002263958786368283, "per_seed": [ 0.09782364219427109, 0.09844150394201279, 0.09902690351009369, 0.09621328115463257, 0.10131707042455673, 0.10156738013029099, 0.09458042681217194, 0.09677895158529282 ], "n": 8 } } ] }, "idea": { "mean": 0.06668837741017342, "std": 0.00810202264282473, "per_seed": [ 0.06714659929275513, 0.07387638092041016, 0.06558588147163391, 0.059248507022857666, 0.05166328698396683, 0.06404716521501541, 0.07355235517024994, 0.07838684320449829 ], "n": 8 }, "comparison": { "delta_mean": -0.030599686317145824, "idea_wins": 8, "n_pairs": 8, "per_seed_diffs": [ -0.02988404780626297, -0.022675395011901855, -0.03340216726064682, -0.0368194580078125, -0.05047975480556488, -0.03556853532791138, -0.018513455986976624, -0.01745467633008957 ], "p_value": 0.0081, "mde": 0.00920648763417379, "mde_rel_pct": 9.463121457507778, "verdict": "idea better (significant)", "system_worked": true }, "custom_track": { "name": "graph_landmark_forecast", "file": "graph_landmark_track.py", "domain": "graph-nn" }, "idea_config": { "lr": 0.01 }, "idea_sweep": [ { "cfg": { "lr": 0.001 }, "result": { "mean": 0.0952951293438673, "std": 0.0034853121866759325, "per_seed": [ 0.0953633040189743, 0.09373979270458221, 0.09905679523944855, 0.092075414955616, 0.10213538259267807, 0.0953550785779953, 0.09400807321071625, 0.09062719345092773 ], "n": 8 } }, { "cfg": { "lr": 0.003 }, "result": { "mean": 0.08661147300153971, "std": 0.004209441400448157, "per_seed": [ 0.08820009231567383, 0.09083075076341629, 0.0886211097240448, 0.08095572143793106, 0.09157451242208481, 0.07902113348245621, 0.0880790650844574, 0.08560939878225327 ], "n": 8 } }, { "cfg": { "lr": 0.01 }, "result": { "mean": 0.06668837741017342, "std": 0.00810202264282473, "per_seed": [ 0.06714659929275513, 0.07387638092041016, 0.06558588147163391, 0.059248507022857666, 0.05166328698396683, 0.06404716521501541, 0.07355235517024994, 0.07838684320449829 ], "n": 8 } } ], "mechanism_signature": { "prediction": "expected collisions are bounded by the hypergeometric union bound", "c_min": 1, "s": 8, "predicted_collision_upper_bound": 114.0, "observed_idea_collision_count": 0.0, "baseline_collision_count": 19.0, "trained_model_observations": [ { "seed": 0, "baseline_metric": 0.0970306470990181, "idea_metric": 0.08820009231567383, "baseline_collisions": 19, "idea_collisions": 0, "idea_pred_target_corr": 0.892614061575404 }, { "seed": 1, "baseline_metric": 0.09655177593231201, "idea_metric": 0.09083075076341629, "baseline_collisions": 19, "idea_collisions": 0, "idea_pred_target_corr": 0.886642980654682 }, { "seed": 2, "baseline_metric": 0.09898804873228073, "idea_metric": 0.0886211097240448, "baseline_collisions": 19, "idea_collisions": 0, "idea_pred_target_corr": 0.8899489630055591 }, { "seed": 3, "baseline_metric": 0.09606796503067017, "idea_metric": 0.08095572143793106, "baseline_collisions": 19, "idea_collisions": 0, "idea_pred_target_corr": 0.8968686834235331 }, { "seed": 4, "baseline_metric": 0.10214304178953171, "idea_metric": 0.09157451242208481, "baseline_collisions": 19, "idea_collisions": 0, "idea_pred_target_corr": 0.8833831997989209 }, { "seed": 5, "baseline_metric": 0.09961570054292679, "idea_metric": 0.07902113348245621, "baseline_collisions": 19, "idea_collisions": 0, "idea_pred_target_corr": 0.8995094696825414 }, { "seed": 6, "baseline_metric": 0.09206581115722656, "idea_metric": 0.0880790650844574, "baseline_collisions": 19, "idea_collisions": 0, "idea_pred_target_corr": 0.8922238907368641 }, { "seed": 7, "baseline_metric": 0.09584151953458786, "idea_metric": 0.08560939878225327, "baseline_collisions": 19, "idea_collisions": 0, "idea_pred_target_corr": 0.8906957622074757 } ], "confirmed": true } }