{ "track": "local_tabular_low_rank_regression", "protocol_status": "fixed bench unavailable at /home/maxwelhelp/all/math2nn/bench; this is supplemental only", "baseline_sweep": [ { "lr": 0.001, "baseline_mean": 0.3229888416826725, "idea_mean": 0.32168807089328766, "delta_mean": -0.001300770789384842, "p_value": 0.2459 }, { "lr": 0.003, "baseline_mean": 0.30325367487967014, "idea_mean": 0.3026796840131283, "delta_mean": -0.0005739908665418625, "p_value": 0.2211 }, { "lr": 0.006, "baseline_mean": 0.3067930955439806, "idea_mean": 0.3066288847476244, "delta_mean": -0.00016421079635620117, "p_value": 0.375 } ], "best_baseline": { "lr": 0.003, "baseline_mean": 0.30325367487967014, "idea_mean": 0.3026796840131283, "delta_mean": -0.0005739908665418625, "p_value": 0.2211 }, "best_idea": { "lr": 0.003, "baseline_mean": 0.30325367487967014, "idea_mean": 0.3026796840131283, "delta_mean": -0.0005739908665418625, "p_value": 0.2211 }, "paired_results": [ { "seed": 0, "lr": 0.001, "baseline": { "mse": 0.29650869965553284, "predicted_retained_proxy": 0.1937541961669922, "observed_retained_response": 0.5846818760037422, "rank": 6 }, "idea": { "mse": 0.29356062412261963, "predicted_retained_proxy": 0.15797016024589539, "observed_retained_response": 0.5918992236256599, "rank": 6 }, "delta": -0.002948075532913208 }, { "seed": 1, "lr": 0.001, "baseline": { "mse": 0.2703535556793213, "predicted_retained_proxy": 0.23858273029327393, "observed_retained_response": 0.5590907707810402, "rank": 6 }, "idea": { "mse": 0.2698759138584137, "predicted_retained_proxy": 0.23667477071285248, "observed_retained_response": 0.5697524696588516, "rank": 6 }, "delta": -0.0004776418209075928 }, { "seed": 2, "lr": 0.001, "baseline": { "mse": 0.3012450337409973, "predicted_retained_proxy": 0.16559083759784698, "observed_retained_response": 0.564979299902916, "rank": 6 }, "idea": { "mse": 0.290658175945282, "predicted_retained_proxy": 0.11861691623926163, "observed_retained_response": 0.5777160674333572, "rank": 6 }, "delta": -0.010586857795715332 }, { "seed": 3, "lr": 0.001, "baseline": { "mse": 0.30093708634376526, "predicted_retained_proxy": 0.13248607516288757, "observed_retained_response": 0.6169104352593422, "rank": 6 }, "idea": { "mse": 0.3020533621311188, "predicted_retained_proxy": 0.09665870666503906, "observed_retained_response": 0.6377759277820587, "rank": 6 }, "delta": 0.0011162757873535156 }, { "seed": 4, "lr": 0.001, "baseline": { "mse": 0.2939927875995636, "predicted_retained_proxy": 0.1950964331626892, "observed_retained_response": 0.5940887182950974, "rank": 6 }, "idea": { "mse": 0.29432255029678345, "predicted_retained_proxy": 0.18268471956253052, "observed_retained_response": 0.6008488908410072, "rank": 6 }, "delta": 0.00032976269721984863 }, { "seed": 5, "lr": 0.001, "baseline": { "mse": 0.5648365020751953, "predicted_retained_proxy": 0.21797959506511688, "observed_retained_response": 0.47909506037831306, "rank": 6 }, "idea": { "mse": 0.5733405351638794, "predicted_retained_proxy": 0.1479392945766449, "observed_retained_response": 0.4866265542805195, "rank": 6 }, "delta": 0.008504033088684082 }, { "seed": 6, "lr": 0.001, "baseline": { "mse": 0.32254257798194885, "predicted_retained_proxy": 0.23743118345737457, "observed_retained_response": 0.5972476936876774, "rank": 6 }, "idea": { "mse": 0.318254679441452, "predicted_retained_proxy": 0.18390978872776031, "observed_retained_response": 0.6131080314517021, "rank": 6 }, "delta": -0.004287898540496826 }, { "seed": 7, "lr": 0.001, "baseline": { "mse": 0.23349449038505554, "predicted_retained_proxy": 0.1556420624256134, "observed_retained_response": 0.5693437606096268, "rank": 6 }, "idea": { "mse": 0.23143872618675232, "predicted_retained_proxy": 0.13866308331489563, "observed_retained_response": 0.5724697560071945, "rank": 6 }, "delta": -0.0020557641983032227 }, { "seed": 0, "lr": 0.003, "baseline": { "mse": 0.28676044940948486, "predicted_retained_proxy": 0.08567658811807632, "observed_retained_response": 0.6662804409861565, "rank": 6 }, "idea": { "mse": 0.28411293029785156, "predicted_retained_proxy": 0.055248551070690155, "observed_retained_response": 0.6709468439221382, "rank": 6 }, "delta": -0.0026475191116333008 }, { "seed": 1, "lr": 0.003, "baseline": { "mse": 0.24638411402702332, "predicted_retained_proxy": 0.15895552933216095, "observed_retained_response": 0.6180978156626225, "rank": 6 }, "idea": { "mse": 0.24585376679897308, "predicted_retained_proxy": 0.07616078108549118, "observed_retained_response": 0.5821574740111828, "rank": 6 }, "delta": -0.0005303472280502319 }, { "seed": 2, "lr": 0.003, "baseline": { "mse": 0.3136810064315796, "predicted_retained_proxy": 0.3495432734489441, "observed_retained_response": 0.6892230659723282, "rank": 6 }, "idea": { "mse": 0.3149034380912781, "predicted_retained_proxy": 0.196433886885643, "observed_retained_response": 0.6687191724777222, "rank": 6 }, "delta": 0.0012224316596984863 }, { "seed": 3, "lr": 0.003, "baseline": { "mse": 0.2992812395095825, "predicted_retained_proxy": 0.11782639473676682, "observed_retained_response": 0.6399654373526573, "rank": 6 }, "idea": { "mse": 0.2992812395095825, "predicted_retained_proxy": 0.11782639473676682, "observed_retained_response": 0.6399654373526573, "rank": 6 }, "delta": 0.0 }, { "seed": 4, "lr": 0.003, "baseline": { "mse": 0.27185118198394775, "predicted_retained_proxy": 0.11754812300205231, "observed_retained_response": 0.6420586034655571, "rank": 6 }, "idea": { "mse": 0.27225157618522644, "predicted_retained_proxy": 0.10567549616098404, "observed_retained_response": 0.6364488303661346, "rank": 6 }, "delta": 0.0004003942012786865 }, { "seed": 5, "lr": 0.003, "baseline": { "mse": 0.5052655935287476, "predicted_retained_proxy": 0.17139050364494324, "observed_retained_response": 0.6117693334817886, "rank": 6 }, "idea": { "mse": 0.5055328011512756, "predicted_retained_proxy": 0.15541502833366394, "observed_retained_response": 0.6354210749268532, "rank": 6 }, "delta": 0.00026720762252807617 }, { "seed": 6, "lr": 0.003, "baseline": { "mse": 0.2950036823749542, "predicted_retained_proxy": 0.15834441781044006, "observed_retained_response": 0.6878739595413208, "rank": 6 }, "idea": { "mse": 0.2912462651729584, "predicted_retained_proxy": 0.13159003853797913, "observed_retained_response": 0.6799127534031868, "rank": 6 }, "delta": -0.0037574172019958496 }, { "seed": 7, "lr": 0.003, "baseline": { "mse": 0.20780213177204132, "predicted_retained_proxy": 0.07717398554086685, "observed_retained_response": 0.611984059214592, "rank": 6 }, "idea": { "mse": 0.20825545489788055, "predicted_retained_proxy": 0.05258169770240784, "observed_retained_response": 0.5936216339468956, "rank": 6 }, "delta": 0.0004533231258392334 }, { "seed": 0, "lr": 0.006, "baseline": { "mse": 0.2900365889072418, "predicted_retained_proxy": 0.08479925990104675, "observed_retained_response": 0.6431076973676682, "rank": 6 }, "idea": { "mse": 0.2889346182346344, "predicted_retained_proxy": 0.07691498100757599, "observed_retained_response": 0.6423569396138191, "rank": 6 }, "delta": -0.0011019706726074219 }, { "seed": 1, "lr": 0.006, "baseline": { "mse": 0.2498152107000351, "predicted_retained_proxy": 0.16318391263484955, "observed_retained_response": 0.6058170236647129, "rank": 6 }, "idea": { "mse": 0.2503966987133026, "predicted_retained_proxy": 0.15665803849697113, "observed_retained_response": 0.6152011044323444, "rank": 6 }, "delta": 0.0005814880132675171 }, { "seed": 2, "lr": 0.006, "baseline": { "mse": 0.3184656798839569, "predicted_retained_proxy": 0.34392213821411133, "observed_retained_response": 0.6691530644893646, "rank": 6 }, "idea": { "mse": 0.3190155625343323, "predicted_retained_proxy": 0.18742811679840088, "observed_retained_response": 0.6492236107587814, "rank": 6 }, "delta": 0.0005498826503753662 }, { "seed": 3, "lr": 0.006, "baseline": { "mse": 0.3039996325969696, "predicted_retained_proxy": 0.10392030328512192, "observed_retained_response": 0.6193787045776844, "rank": 6 }, "idea": { "mse": 0.3013850748538971, "predicted_retained_proxy": 0.10185642540454865, "observed_retained_response": 0.6289432123303413, "rank": 6 }, "delta": -0.0026145577430725098 }, { "seed": 4, "lr": 0.006, "baseline": { "mse": 0.27291032671928406, "predicted_retained_proxy": 0.11452590674161911, "observed_retained_response": 0.6283508837223053, "rank": 6 }, "idea": { "mse": 0.27344390749931335, "predicted_retained_proxy": 0.10965780168771744, "observed_retained_response": 0.621874064207077, "rank": 6 }, "delta": 0.0005335807800292969 }, { "seed": 5, "lr": 0.006, "baseline": { "mse": 0.5082470774650574, "predicted_retained_proxy": 0.1677103191614151, "observed_retained_response": 0.577905185520649, "rank": 6 }, "idea": { "mse": 0.5096269249916077, "predicted_retained_proxy": 0.1527785360813141, "observed_retained_response": 0.6050795763731003, "rank": 6 }, "delta": 0.001379847526550293 }, { "seed": 6, "lr": 0.006, "baseline": { "mse": 0.29952022433280945, "predicted_retained_proxy": 0.15662352740764618, "observed_retained_response": 0.6685446798801422, "rank": 6 }, "idea": { "mse": 0.29916277527809143, "predicted_retained_proxy": 0.1364256590604782, "observed_retained_response": 0.6665552407503128, "rank": 6 }, "delta": -0.0003574490547180176 }, { "seed": 7, "lr": 0.006, "baseline": { "mse": 0.21135002374649048, "predicted_retained_proxy": 0.0768255814909935, "observed_retained_response": 0.5843467249069363, "rank": 6 }, "idea": { "mse": 0.21106551587581635, "predicted_retained_proxy": 0.06295523047447205, "observed_retained_response": 0.59474778175354, "rank": 6 }, "delta": -0.0002845078706741333 } ], "bench_report": { "baseline_sweep": [ { "lr": 0.001, "baseline_mean": 0.3229888416826725, "idea_mean": 0.32168807089328766, "delta_mean": -0.001300770789384842, "p_value": 0.2459 }, { "lr": 0.003, "baseline_mean": 0.30325367487967014, "idea_mean": 0.3026796840131283, "delta_mean": -0.0005739908665418625, "p_value": 0.2211 }, { "lr": 0.006, "baseline_mean": 0.3067930955439806, "idea_mean": 0.3066288847476244, "delta_mean": -0.00016421079635620117, "p_value": 0.375 } ], "idea_per_seed_results": [ { "seed": 0, "lr": 0.003, "baseline": { "mse": 0.28676044940948486, "predicted_retained_proxy": 0.08567658811807632, "observed_retained_response": 0.6662804409861565, "rank": 6 }, "idea": { "mse": 0.28411293029785156, "predicted_retained_proxy": 0.055248551070690155, "observed_retained_response": 0.6709468439221382, "rank": 6 }, "delta": -0.0026475191116333008 }, { "seed": 1, "lr": 0.003, "baseline": { "mse": 0.24638411402702332, "predicted_retained_proxy": 0.15895552933216095, "observed_retained_response": 0.6180978156626225, "rank": 6 }, "idea": { "mse": 0.24585376679897308, "predicted_retained_proxy": 0.07616078108549118, "observed_retained_response": 0.5821574740111828, "rank": 6 }, "delta": -0.0005303472280502319 }, { "seed": 2, "lr": 0.003, "baseline": { "mse": 0.3136810064315796, "predicted_retained_proxy": 0.3495432734489441, "observed_retained_response": 0.6892230659723282, "rank": 6 }, "idea": { "mse": 0.3149034380912781, "predicted_retained_proxy": 0.196433886885643, "observed_retained_response": 0.6687191724777222, "rank": 6 }, "delta": 0.0012224316596984863 }, { "seed": 3, "lr": 0.003, "baseline": { "mse": 0.2992812395095825, "predicted_retained_proxy": 0.11782639473676682, "observed_retained_response": 0.6399654373526573, "rank": 6 }, "idea": { "mse": 0.2992812395095825, "predicted_retained_proxy": 0.11782639473676682, "observed_retained_response": 0.6399654373526573, "rank": 6 }, "delta": 0.0 }, { "seed": 4, "lr": 0.003, "baseline": { "mse": 0.27185118198394775, "predicted_retained_proxy": 0.11754812300205231, "observed_retained_response": 0.6420586034655571, "rank": 6 }, "idea": { "mse": 0.27225157618522644, "predicted_retained_proxy": 0.10567549616098404, "observed_retained_response": 0.6364488303661346, "rank": 6 }, "delta": 0.0004003942012786865 }, { "seed": 5, "lr": 0.003, "baseline": { "mse": 0.5052655935287476, "predicted_retained_proxy": 0.17139050364494324, "observed_retained_response": 0.6117693334817886, "rank": 6 }, "idea": { "mse": 0.5055328011512756, "predicted_retained_proxy": 0.15541502833366394, "observed_retained_response": 0.6354210749268532, "rank": 6 }, "delta": 0.00026720762252807617 }, { "seed": 6, "lr": 0.003, "baseline": { "mse": 0.2950036823749542, "predicted_retained_proxy": 0.15834441781044006, "observed_retained_response": 0.6878739595413208, "rank": 6 }, "idea": { "mse": 0.2912462651729584, "predicted_retained_proxy": 0.13159003853797913, "observed_retained_response": 0.6799127534031868, "rank": 6 }, "delta": -0.0037574172019958496 }, { "seed": 7, "lr": 0.003, "baseline": { "mse": 0.20780213177204132, "predicted_retained_proxy": 0.07717398554086685, "observed_retained_response": 0.611984059214592, "rank": 6 }, "idea": { "mse": 0.20825545489788055, "predicted_retained_proxy": 0.05258169770240784, "observed_retained_response": 0.5936216339468956, "rank": 6 }, "delta": 0.0004533231258392334 } ], "paired_delta_mean": -0.0005739908665418625, "permutation_p_value": 0.2211 }, "mechanism_signature": { "predicted": "retained rank-one first-order contribution proxy", "observed": "mean absolute retained rank-one response on held-out inputs", "predicted_values": [ 0.15797016024589539, 0.23667477071285248, 0.11861691623926163, 0.09665870666503906, 0.18268471956253052, 0.1479392945766449, 0.18390978872776031, 0.13866308331489563, 0.055248551070690155, 0.07616078108549118, 0.196433886885643, 0.11782639473676682, 0.10567549616098404, 0.15541502833366394, 0.13159003853797913, 0.05258169770240784, 0.07691498100757599, 0.15665803849697113, 0.18742811679840088, 0.10185642540454865, 0.10965780168771744, 0.1527785360813141, 0.1364256590604782, 0.06295523047447205 ], "observed_values": [ 0.5918992236256599, 0.5697524696588516, 0.5777160674333572, 0.6377759277820587, 0.6008488908410072, 0.4866265542805195, 0.6131080314517021, 0.5724697560071945, 0.6709468439221382, 0.5821574740111828, 0.6687191724777222, 0.6399654373526573, 0.6364488303661346, 0.6354210749268532, 0.6799127534031868, 0.5936216339468956, 0.6423569396138191, 0.6152011044323444, 0.6492236107587814, 0.6289432123303413, 0.621874064207077, 0.6050795763731003, 0.6665552407503128, 0.59474778175354 ], "confirmed": false } }