{ "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": 1000, "seed": 123, "gpu": 0, "output_dir": "results/toy_lq", "vnet_hidden": 256, "vnet_layers": 3, "term_grad_weight": 1.0, "fm_weight": 0.0 }, "log": { "steps": [ 1, 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000 ], "dfa_costate_cos": [ 0.0061469420325011015, 0.007492478704079986, 0.006436596314112346, 0.001648913836106658, 0.005657717352733016, 0.010142655577510595, 0.005493079700196783, 0.008208209726338586, 0.002802101274331411 ], "state_costate_cos": [ 0.04875442975511154, 0.9345830877621969, 0.9331425180037817, 0.9344809154669443, 0.9360056469837824, 0.9349906196196874, 0.9401087015867233, 0.9377894500891367, 0.9381228536367416 ], "credit_costate_cos": [ 0.005350367398932576, 0.8715064575274786, 0.9082922885815302, 0.9268463253974915, 0.9348577807346979, 0.9338823060194651, 0.9398750513792038, 0.9371217240889868, 0.9403592944145203 ], "dfa_rho": [ 0.014851124413932363, 0.004832483362406492, 0.005500619454930226, 0.0014784028753638268, 0.0024716570042073727, 0.002679668522129456, 0.004042171291075647, 0.007841781480237842, -0.003731151965136329 ], "state_rho": [ 0.0525277191773057, 0.9209283490975698, 0.9212760378917059, 0.9215241422255834, 0.9252830098072687, 0.9173514097929001, 0.9268933484951655, 0.9245105236768723, 0.9267788628737131 ], "credit_rho": [ 8.900166722014546e-05, 0.8210234741369883, 0.8804274102052053, 0.9100336680809656, 0.9228341629107794, 0.9162226617336273, 0.9262114216883978, 0.9239083131154379, 0.9273379941781362 ], "dfa_nudge": [ -0.0020856610499322414, -0.002391135785728693, -0.0018826122395694256, 0.0002794961134592692, -0.0017084906188150246, -0.0027558018919080496, -0.0014085437481602032, -0.0023310642379025617, -0.0009137461893260479 ], "state_nudge": [ -0.01762500188002984, -0.31959830472866696, -0.3143775438268979, -0.3134472444653511, -0.33237058420976, -0.3183835695187251, -0.32320864746967953, -0.32294898976882297, -0.31496328860521317 ], "credit_nudge": [ -0.0004618208234508832, -0.2996478999654452, -0.30627985050280887, -0.3099561383326848, -0.33090290675560635, -0.31673717498779297, -0.3218600004911423, -0.32163529098033905, -0.31442063798507053 ], "bridge_residual": [], "state_bridge_loss": [ 64.66886901855469, 1.9055249691009521, 1.9802964925765991, 1.9747117757797241, 1.9333608150482178, 2.0244460105895996, 2.235288381576538, 1.8483705520629883, 1.833404541015625 ], "credit_bridge_loss": [ 129.2601776123047, 8.826760292053223, 8.817267417907715, 9.7413911819458, 8.333325386047363, 8.390682220458984, 8.381990432739258, 9.069635391235352, 8.738546371459961 ], "term_loss": [ 109.68403625488281, 3.380112648010254, 4.175432205200195, 4.737654685974121, 3.632157802581787, 3.3351938724517822, 3.776655912399292, 4.416824817657471, 4.120006084442139 ], "bridge_loss": [ 5.943464884694549e-07, 0.2433367669582367, 0.18713834881782532, 0.12417592853307724, 0.13950209319591522, 0.14192476868629456, 0.15276893973350525, 0.10663559287786484, 0.12137635797262192 ], "term_grad_loss": [ 19.57614517211914, 5.203310966491699, 4.4546966552734375, 4.879560470581055, 4.5616655349731445, 4.9135637283325195, 4.452565670013428, 4.546175003051758, 4.497163772583008 ], "fm_loss": [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ] }, "final_per_layer": { "dfa_costate_cos": [ 0.04479089006781578, -0.04426354169845581, 0.008874599821865559, 0.05599997937679291, 0.02961653470993042, -0.022058574482798576, 0.027186769992113113, -0.0337681919336319, -0.020245034247636795, -0.04076787084341049, 0.006424968130886555, 0.021834686398506165 ], "state_costate_cos": [ 0.9351104497909546, 0.9362065196037292, 0.9370632171630859, 0.9374789595603943, 0.9381210803985596, 0.9385455846786499, 0.9386894702911377, 0.9390624165534973, 0.9393105506896973, 0.9393242597579956, 0.9392982721328735, 0.9392634630203247 ], "credit_costate_cos": [ 0.9360430240631104, 0.9369803667068481, 0.9380120038986206, 0.9385954141616821, 0.9395070672035217, 0.9403176307678223, 0.9409258961677551, 0.9413450956344604, 0.9420279264450073, 0.9428501725196838, 0.9435971975326538, 0.9441097378730774 ], "dfa_rho": [ 0.041250549256801605, -0.049739208072423935, 0.00025176629424095154, 0.04237007349729538, 0.040798719972372055, -0.037202395498752594, 0.004211327061057091, 0.0017704367637634277, -0.04704931750893593, -0.02992381900548935, -0.01541070081293583, 0.003898744471371174 ], "state_rho": [ 0.9230573177337646, 0.9292065501213074, 0.9277745485305786, 0.926990270614624, 0.930223822593689, 0.9262816905975342, 0.9238564968109131, 0.9235492944717407, 0.927101731300354, 0.929185152053833, 0.9233707189559937, 0.9307487607002258 ], "credit_rho": [ 0.924521803855896, 0.9213208556175232, 0.923781156539917, 0.9222273826599121, 0.9230118989944458, 0.9317750930786133, 0.9238618016242981, 0.9351733326911926, 0.9263075590133667, 0.9302359819412231, 0.9358923435211182, 0.9299467206001282 ], "dfa_nudge": [ -0.01619502529501915, 0.016632290557026863, -0.004560360684990883, -0.018955951556563377, -0.011058392003178596, 0.007517071440815926, -0.008148249238729477, 0.01091383583843708, 0.006811931729316711, 0.015017258003354073, -0.000978708267211914, -0.00796065479516983 ], "state_nudge": [ -0.31427979469299316, -0.3147982358932495, -0.3148774206638336, -0.3150269389152527, -0.31536394357681274, -0.3156891465187073, -0.3149397373199463, -0.3145085573196411, -0.3144562840461731, -0.31494998931884766, -0.3152415156364441, -0.31542789936065674 ], "credit_nudge": [ -0.31250905990600586, -0.3131006360054016, -0.31331712007522583, -0.31366780400276184, -0.3142547011375427, -0.3149237632751465, -0.31442567706108093, -0.314262330532074, -0.31445154547691345, -0.3153681755065918, -0.3161371052265167, -0.3166297376155853 ] } }