From d8fe36a0c44fd45edb6a637d3530fdaeeca74fd5 Mon Sep 17 00:00:00 2001 From: YurenHao0426 Date: Thu, 6 Aug 2026 17:15:51 -0500 Subject: results: audit Rain EP bias confirmation --- results/ep_bias/c1_gate.json | 142 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 142 insertions(+) create mode 100644 results/ep_bias/c1_gate.json (limited to 'results/ep_bias/c1_gate.json') diff --git a/results/ep_bias/c1_gate.json b/results/ep_bias/c1_gate.json new file mode 100644 index 0000000..1266bba --- /dev/null +++ b/results/ep_bias/c1_gate.json @@ -0,0 +1,142 @@ +{ + "author_revision": "6b253fd8a5d267535f58ab79992256ef10031ceb", + "checks": { + "clean_deficit_upper_95_below_3_points": false, + "finite_complete_required_conditions": true, + "innovation_beats_constant_every_seed": false, + "innovation_beats_raw_every_seed": true, + "innovation_constant_lower_95_at_least_5_points": false, + "innovation_raw_lower_95_at_least_10_points": true, + "innovation_residual_below_constant_every_seed": true, + "matched_128_neutral_observations": true, + "mean_clean_validation_at_least_70": true, + "mean_wall_overhead_at_most_15_percent": true, + "noise_abs_mean_upper_95_below_3_points": false, + "oracle_abs_mean_upper_95_below_3_points": false, + "protocol_fixed_and_bp_free": true, + "raw_damage_at_least_10_points_every_seed": true, + "single_author_revision": true, + "single_sdil_revision": true + }, + "gate": "fail", + "mean_clean_validation_accuracy": 0.7626000000000001, + "num_records": 30, + "rows": [ + { + "clean": 0.7705, + "clean_minus_innovation": 0.020499999999999963, + "constant": 0.681, + "innovation": 0.75, + "innovation_minus_constant": 0.06899999999999995, + "innovation_minus_oracle": -0.01749999999999996, + "innovation_minus_raw": 0.5385, + "innovation_wall_over_clean": 1.0200028100178125, + "noise_minus_clean": 0.026000000000000023, + "oracle": 0.7675, + "raw": 0.2115, + "same_rms_noise": 0.7965, + "seed": 1989 + }, + { + "clean": 0.716, + "clean_minus_innovation": -0.020500000000000074, + "constant": 0.7915, + "innovation": 0.7365, + "innovation_minus_constant": -0.05499999999999994, + "innovation_minus_oracle": -0.03849999999999998, + "innovation_minus_raw": 0.44850000000000007, + "innovation_wall_over_clean": 1.002160782870049, + "noise_minus_clean": 0.03949999999999998, + "oracle": 0.775, + "raw": 0.288, + "same_rms_noise": 0.7555, + "seed": 1990 + }, + { + "clean": 0.7485, + "clean_minus_innovation": -0.017999999999999905, + "constant": 0.5425, + "innovation": 0.7665, + "innovation_minus_constant": 0.22399999999999998, + "innovation_minus_oracle": 0.10649999999999993, + "innovation_minus_raw": 0.6705, + "innovation_wall_over_clean": 0.9799086026637867, + "noise_minus_clean": -0.08100000000000007, + "oracle": 0.66, + "raw": 0.096, + "same_rms_noise": 0.6675, + "seed": 1991 + }, + { + "clean": 0.8, + "clean_minus_innovation": 0.03500000000000003, + "constant": 0.7125, + "innovation": 0.765, + "innovation_minus_constant": 0.05249999999999999, + "innovation_minus_oracle": -0.03200000000000003, + "innovation_minus_raw": 0.35100000000000003, + "innovation_wall_over_clean": 0.9760042015047453, + "noise_minus_clean": -0.007500000000000062, + "oracle": 0.797, + "raw": 0.414, + "same_rms_noise": 0.7925, + "seed": 1992 + }, + { + "clean": 0.778, + "clean_minus_innovation": 0.07000000000000006, + "constant": 0.6675, + "innovation": 0.708, + "innovation_minus_constant": 0.04049999999999998, + "innovation_minus_oracle": -0.07050000000000001, + "innovation_minus_raw": 0.14849999999999997, + "innovation_wall_over_clean": 1.0005652695437517, + "noise_minus_clean": -0.018000000000000016, + "oracle": 0.7785, + "raw": 0.5595, + "same_rms_noise": 0.76, + "seed": 1993 + } + ], + "sdil_revision": "a10a84066a6f7aa38d86b1a71ed371c9c6f22815", + "stage": "rain_ep_bias_c1", + "statistics": { + "clean_minus_innovation": { + "lower_95": -0.018827912063523824, + "mean": 0.017400000000000016, + "sem": 0.016993675294061622, + "upper_95": 0.05362791206352385 + }, + "innovation_minus_constant": { + "lower_95": -0.029669484192664783, + "mean": 0.0662, + "sem": 0.044970156770907516, + "upper_95": 0.16206948419266476 + }, + "innovation_minus_oracle": { + "lower_95": -0.0753849705784051, + "mean": -0.01040000000000001, + "sem": 0.030482946051850025, + "upper_95": 0.0753849705784051 + }, + "innovation_minus_raw": { + "lower_95": 0.24346576516238344, + "mean": 0.4314, + "sem": 0.08815560106992634, + "upper_95": 0.6193342348376165 + }, + "innovation_wall_over_clean": { + "lower_95": 0.97858499866313, + "mean": 0.9957283333200291, + "sem": 0.008041541618681968, + "upper_95": 1.0128716679769283 + }, + "noise_minus_clean": { + "lower_95": -0.053022807458936866, + "mean": -0.00820000000000003, + "sem": 0.02102534185215547, + "upper_95": 0.053022807458936866 + } + }, + "test_policy": "test_set_never_loaded; fixed epoch-3 train holdout" +} -- cgit v1.2.3