{ "config": { "d_hidden": 64, "output_dim": 10, "num_layers": 12, "sigma": 0.03, "batch_size": 256, "num_steps": 8000, "lr_fb": 0.001, "lam": 0.1, "K": 8, "ema_momentum": 0.995, "sigma_bridge": 0.1, "eval_every": 2000, "seed": 42, "gpu": 0, "output_dir": "results/toy_lq_frozen", "vnet_hidden": 256, "vnet_layers": 3, "term_grad_weight": 1.0, "fm_weight": 0.0 }, "log": { "steps": [ 1, 2000, 4000, 6000, 8000 ], "dfa_costate_cos": [ 0.001022340264171362, 0.001144224622597297, 0.0035498210539420447, 0.0025704174380128584, 0.0012041970621794462 ], "state_costate_cos": [ 0.009041048353537917, 0.9367498010396957, 0.9443944742282232, 0.9488719999790192, 0.9465398589769999 ], "credit_costate_cos": [ 0.024892715892444055, 0.9009914000829061, 0.9385832945505778, 0.947337324420611, 0.9453924099604288 ], "dfa_rho": [ 0.015879416760678094, 0.003749566695963343, 0.0025427089616035423, 0.0041017483454197645, -0.006658251901778082 ], "state_rho": [ 0.0029661240599428615, 0.9205784201622009, 0.9331746151049932, 0.9367670218149821, 0.9348721752564112 ], "credit_rho": [ 0.02234963719577839, 0.8549301077922186, 0.9257398645083109, 0.9335627655188242, 0.9330280274152756 ], "dfa_nudge": [ -0.0003799900102118651, -0.0001569713931530714, -0.0012285423775513966, -0.00047482581188281375, -0.0003635381193210681 ], "state_nudge": [ -0.002246703254058957, -0.3365800951917966, -0.3378218387564023, -0.34102949251731235, -0.3362140009800593 ], "credit_nudge": [ -0.0079942528779308, -0.32416996111472446, -0.3345658630132675, -0.3391987532377243, -0.33437975496053696 ], "bridge_residual": [], "state_bridge_loss": [ 1.0073249340057373, 0.03208087012171745, 0.02979956567287445, 0.03251900523900986, 0.03256681188941002 ], "credit_bridge_loss": [ 132.09298706054688, 10.374980926513672, 10.61994743347168, 10.229816436767578, 9.931343078613281 ], "term_loss": [ 111.63633728027344, 5.137801647186279, 5.388574600219727, 5.321071147918701, 4.950850486755371 ], "bridge_loss": [ 6.45359421014291e-07, 0.2699485719203949, 0.12590564787387848, 0.13655179738998413, 0.15239471197128296 ], "term_grad_loss": [ 20.456655502319336, 4.967230796813965, 5.105466842651367, 4.772193431854248, 4.828097343444824 ], "fm_loss": [ 0.0, 0.0, 0.0, 0.0, 0.0 ] }, "final_per_layer": { "dfa_costate_cos": [ -0.053928110748529434, -0.03558344021439552, -0.002497725188732147, -0.025605209171772003, -0.0015886849723756313, -0.004757361486554146, 0.038880251348018646, 0.04505593329668045, 0.044440463185310364, 0.049856118857860565, -0.036305028945207596, -0.0035168412141501904 ], "state_costate_cos": [ 0.943811297416687, 0.9446508884429932, 0.9451277256011963, 0.9456416368484497, 0.9462511539459229, 0.9469271898269653, 0.9473466873168945, 0.947996973991394, 0.9481196999549866, 0.9477524757385254, 0.9474242925643921, 0.947428286075592 ], "credit_costate_cos": [ 0.941976010799408, 0.9424219727516174, 0.94321209192276, 0.9437947273254395, 0.9448409080505371, 0.9454640746116638, 0.9459144473075867, 0.9465770721435547, 0.9470815062522888, 0.9474484920501709, 0.947708010673523, 0.9482696056365967 ], "dfa_rho": [ -0.04425683990120888, -0.06072389334440231, -0.02133849635720253, -0.009745828807353973, 0.012875140644609928, -0.06230748072266579, -0.011334436014294624, 0.06567166745662689, 0.05127769336104393, 0.02585187926888466, -0.02806694805622101, 0.0021985196508467197 ], "state_rho": [ 0.9318734407424927, 0.9358463883399963, 0.9371962547302246, 0.932350754737854, 0.9360877871513367, 0.9400110840797424, 0.9344450235366821, 0.9253479242324829, 0.9369418621063232, 0.9354310035705566, 0.9379286170005798, 0.9350059628486633 ], "credit_rho": [ 0.925601065158844, 0.9282341599464417, 0.9294254779815674, 0.9282456040382385, 0.9352800846099854, 0.9381647109985352, 0.9306818246841431, 0.9283095598220825, 0.9359391331672668, 0.9364811182022095, 0.940323531627655, 0.9396500587463379 ], "dfa_nudge": [ 0.018354831263422966, 0.012472891248762608, 0.0010786037892103195, 0.008631331846117973, -0.000161781907081604, 0.0025226473808288574, -0.013589534908533096, -0.017021682113409042, -0.01592499390244484, -0.017948923632502556, 0.015213390812277794, 0.0020107626914978027 ], "state_nudge": [ -0.3381166160106659, -0.33782684803009033, -0.33681023120880127, -0.3365652561187744, -0.3364033102989197, -0.336093544960022, -0.33575671911239624, -0.3358075022697449, -0.33578699827194214, -0.3356783986091614, -0.3347625136375427, -0.33496007323265076 ], "credit_nudge": [ -0.3348296880722046, -0.33472180366516113, -0.33402615785598755, -0.3340165615081787, -0.33416393399238586, -0.33410122990608215, -0.33391931653022766, -0.3341727554798126, -0.334547221660614, -0.33483123779296875, -0.3342774510383606, -0.33494970202445984 ] } }