{ "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": 123, "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": [ 0.0061469420325011015, 0.005306758529817064, 0.0032278880632172027, 0.010428216774016619, 0.005160536073769133 ], "state_costate_cos": [ 0.04923563698927561, 0.9359788099924723, 0.93496473133564, 0.9391002754370371, 0.9422274927298228 ], "credit_costate_cos": [ 0.005350367398932576, 0.9094806561867396, 0.933747818072637, 0.939527784784635, 0.9428367614746094 ], "dfa_rho": [ 0.014851124413932363, -0.0009741405956447124, 0.010637809444839755, 0.01579231140203774, 0.0006830055887500445 ], "state_rho": [ 0.05303209243963162, 0.9233627518018087, 0.9250594973564148, 0.9248303671677908, 0.9291473726431528 ], "credit_rho": [ 8.900166722014546e-05, 0.8767146865526835, 0.9206246634324392, 0.9250840544700623, 0.9295186003049215 ], "dfa_nudge": [ -0.0020856610499322414, -0.0016959290175388257, -0.0004252120852470398, -0.00336669214690725, -0.0014625868449608486 ], "state_nudge": [ -0.017791483240822952, -0.32088013738393784, -0.3230740974346797, -0.318187120060126, -0.324822299182415 ], "credit_nudge": [ -0.0004618208234508832, -0.31195973853270215, -0.3216395452618599, -0.3171723609169324, -0.3237730637192726 ], "bridge_residual": [], "state_bridge_loss": [ 1.0102367401123047, 0.029022935777902603, 0.031384147703647614, 0.02959899604320526, 0.031896352767944336 ], "credit_bridge_loss": [ 129.2601776123047, 8.807141304016113, 8.771026611328125, 7.9213643074035645, 8.751266479492188 ], "term_loss": [ 109.68403625488281, 4.207324028015137, 4.038488388061523, 3.279881477355957, 3.9708642959594727 ], "bridge_loss": [ 5.943464884694549e-07, 0.11741068214178085, 0.08691056072711945, 0.10882671922445297, 0.12579748034477234 ], "term_grad_loss": [ 19.57614517211914, 4.4824066162109375, 4.645627975463867, 4.532656192779541, 4.654605388641357 ], "fm_loss": [ 0.0, 0.0, 0.0, 0.0, 0.0 ] }, "final_per_layer": { "dfa_costate_cos": [ 0.04813992977142334, -0.042106956243515015, 0.004819205962121487, 0.06406223028898239, 0.027982598170638084, -0.024616792798042297, 0.03184102475643158, -0.03284747526049614, -0.031094228848814964, -0.04093015938997269, 0.012475000694394112, 0.0442020557820797 ], "state_costate_cos": [ 0.9390542507171631, 0.9402485489845276, 0.9412673711776733, 0.9416499137878418, 0.9424928426742554, 0.9428901672363281, 0.9427830576896667, 0.9428685903549194, 0.9430990219116211, 0.9434692859649658, 0.9433966875076294, 0.9435101747512817 ], "credit_costate_cos": [ 0.9391632676124573, 0.9400047659873962, 0.9408236742019653, 0.9413831233978271, 0.9420697689056396, 0.942909836769104, 0.9431338906288147, 0.943551778793335, 0.9441136121749878, 0.9448702335357666, 0.9456644654273987, 0.9463527202606201 ], "dfa_rho": [ 0.0134794432669878, -0.042476922273635864, 0.04209040105342865, 0.06388744711875916, -0.012325622141361237, -0.025234051048755646, -0.020182523876428604, -0.02071167342364788, -0.049487367272377014, -0.05043957754969597, 0.04616815596818924, 0.0634283572435379 ], "state_rho": [ 0.9241164922714233, 0.9311657547950745, 0.926814079284668, 0.9336568117141724, 0.9266120791435242, 0.9352933764457703, 0.9264969825744629, 0.9290511012077332, 0.9249886870384216, 0.9313560128211975, 0.9301149845123291, 0.9301021099090576 ], "credit_rho": [ 0.9276995658874512, 0.9273138046264648, 0.9248120784759521, 0.9268039464950562, 0.92667555809021, 0.9278458952903748, 0.9272637367248535, 0.9345529079437256, 0.9366906881332397, 0.9292744398117065, 0.9313026070594788, 0.9339879751205444 ], "dfa_nudge": [ -0.017539501190185547, 0.01593128778040409, -0.0030720066279172897, -0.020152313634753227, -0.009087346494197845, 0.007799159735441208, -0.010878749191761017, 0.013225046917796135, 0.011669810861349106, 0.017475932836532593, -0.006174879148602486, -0.016747483983635902 ], "state_nudge": [ -0.32301613688468933, -0.3241724371910095, -0.32460731267929077, -0.32475656270980835, -0.325019896030426, -0.3253987431526184, -0.32468554377555847, -0.3243948817253113, -0.3248230516910553, -0.32554829120635986, -0.3256106674671173, -0.3258340656757355 ], "credit_nudge": [ -0.32093775272369385, -0.32208389043807983, -0.32268720865249634, -0.3229762613773346, -0.3233621120452881, -0.324078768491745, -0.32360684871673584, -0.32358717918395996, -0.3243335783481598, -0.32536792755126953, -0.3258592486381531, -0.32639598846435547 ] } }