{ "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": 456, "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": [ -0.008305357536301017, -0.008392414000506202, -0.0038773335108999163, -0.005692236203079422, -0.0027285031198213496 ], "state_costate_cos": [ 0.010795300283158818, 0.9460556507110596, 0.9393202712138494, 0.9483461380004883, 0.9449077347914377 ], "credit_costate_cos": [ 0.010942678588132063, 0.9265013436476389, 0.9366317639748255, 0.947089821100235, 0.9429272959629694 ], "dfa_rho": [ -0.0028248391657446823, -0.0069896553953488665, 0.006519075793524583, -0.0016956791902581851, 0.009361718820097545 ], "state_rho": [ 0.02673721546307206, 0.9346788177887598, 0.9264062345027924, 0.9364438454310099, 0.928318440914154 ], "credit_rho": [ 0.015292729716748, 0.8930107355117798, 0.9212059676647186, 0.9316875636577606, 0.9271238495906194 ], "dfa_nudge": [ 0.004384364855165283, 0.004451872392868002, 0.0023875126304725804, 0.0029402777242163816, 0.0018359140182534854 ], "state_nudge": [ -0.005124294043829043, -0.34853988885879517, -0.36369936913251877, -0.3535121629635493, -0.3701842874288559 ], "credit_nudge": [ -0.003232262640570601, -0.3407190019885699, -0.36137687663237256, -0.35145895431439084, -0.3679164672891299 ], "bridge_residual": [], "state_bridge_loss": [ 1.0078721046447754, 0.029677797108888626, 0.03204238414764404, 0.0327574647963047, 0.033611781895160675 ], "credit_bridge_loss": [ 158.73072814941406, 11.051497459411621, 9.754555702209473, 10.701624870300293, 9.23304557800293 ], "term_loss": [ 132.93673706054688, 5.0680437088012695, 4.583424091339111, 5.3644843101501465, 4.062508583068848 ], "bridge_loss": [ 7.166463547036983e-07, 0.25247713923454285, 0.14709019660949707, 0.1511959433555603, 0.12251871824264526 ], "term_grad_loss": [ 25.793991088867188, 5.730976581573486, 5.024041175842285, 5.185944557189941, 5.048018932342529 ], "fm_loss": [ 0.0, 0.0, 0.0, 0.0, 0.0 ] }, "final_per_layer": { "dfa_costate_cos": [ 0.005115414969623089, -0.0511106476187706, 0.026345418766140938, -0.06129831820726395, 0.023028161376714706, 0.043917812407016754, -0.02071939967572689, 0.020611021667718887, -0.025742489844560623, -0.052854984998703, 0.004462959244847298, 0.05550301447510719 ], "state_costate_cos": [ 0.9431805610656738, 0.9437761306762695, 0.9443435072898865, 0.9446961879730225, 0.9448176622390747, 0.9449890851974487, 0.945353627204895, 0.9453505277633667, 0.9454029202461243, 0.9456251859664917, 0.945793867111206, 0.945563554763794 ], "credit_costate_cos": [ 0.9382357001304626, 0.9392683506011963, 0.9404538869857788, 0.9414603114128113, 0.9420560598373413, 0.9428950548171997, 0.9436970353126526, 0.9444013833999634, 0.9445676803588867, 0.9454110860824585, 0.9460842609405518, 0.9465967416763306 ], "dfa_rho": [ 0.0005743983201682568, -0.06480635702610016, 0.04855071008205414, -0.09910713136196136, 0.0678277313709259, 0.07211608439683914, -0.024130607023835182, 0.001894976943731308, -0.009040179662406445, -0.02709592692553997, 0.047522202134132385, 0.09803472459316254 ], "state_rho": [ 0.9245786666870117, 0.9240697026252747, 0.9293171167373657, 0.9301248788833618, 0.9324039220809937, 0.9262233376502991, 0.9288182854652405, 0.9328013062477112, 0.926672101020813, 0.928871750831604, 0.9281368255615234, 0.9278033971786499 ], "credit_rho": [ 0.9264800548553467, 0.9184226989746094, 0.9257187247276306, 0.9245353937149048, 0.9274380803108215, 0.9219262599945068, 0.9251310229301453, 0.9257712960243225, 0.9327481985092163, 0.9326720833778381, 0.9336825013160706, 0.9309598803520203 ], "dfa_nudge": [ -0.0008005127310752869, 0.01917375810444355, -0.009830990806221962, 0.02491075173020363, -0.007071066647768021, -0.017435496672987938, 0.009654557332396507, -0.005623402073979378, 0.01048743724822998, 0.022874275222420692, -0.0009297188371419907, -0.02337862364947796 ], "state_nudge": [ -0.3704739809036255, -0.3704678416252136, -0.36917200684547424, -0.36994534730911255, -0.3694732189178467, -0.37023210525512695, -0.3705016076564789, -0.37048155069351196, -0.36992865800857544, -0.37050265073776245, -0.37076336145401, -0.37026911973953247 ], "credit_nudge": [ -0.3661743402481079, -0.3663756549358368, -0.3655211329460144, -0.3667157292366028, -0.3666227161884308, -0.36782804131507874, -0.36846989393234253, -0.3688392639160156, -0.3685477077960968, -0.36957746744155884, -0.37019991874694824, -0.37012574076652527 ] } }