{ "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": 500, "seed": 42, "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, 500, 1000, 1500, 2000, 2500, 3000, 3500, 4000, 4500, 5000, 5500, 6000, 6500, 7000, 7500, 8000 ], "dfa_costate_cos": [ 0.001022340264171362, 0.00324871806272616, 0.003955680275491129, -6.335701133745412e-05, 0.0009047117297692845, -0.0015635235855976741, 0.0050255006256823736, 0.003974012487257521, -0.0024126653637116155, 0.0010036304011009634, -0.0006338046708454689, 0.00017608713824301958, -9.714129070440929e-05, -0.0011926805212472875, -0.0026168684319903455, -0.003671417556082209, 0.0016770829679444432 ], "state_costate_cos": [ 0.009337988216429949, 0.943408285578092, 0.9517824550469717, 0.9448912839094797, 0.9432462950547537, 0.9444176405668259, 0.9435405482848486, 0.9443136354287466, 0.9458288550376892, 0.9465046127637228, 0.9465998113155365, 0.9461832990248998, 0.9424780905246735, 0.9474884221951166, 0.9449383119742075, 0.9425351719061533, 0.9405365288257599 ], "credit_costate_cos": [ 0.024892715892444055, 0.8234576731920242, 0.8861371924479803, 0.8940252065658569, 0.9119344304005305, 0.928387979666392, 0.9261951943238577, 0.9374307443698248, 0.939670259753863, 0.9423713783423106, 0.9447299987077713, 0.944857731461525, 0.9409955541292826, 0.9459459533294042, 0.9439971546332041, 0.9420813073714575, 0.9400773843129476 ], "dfa_rho": [ 0.015879416760678094, 0.004265802912414074, 0.008714484671751658, 0.001407407767449816, 0.007156838779337704, -0.0010706717148423195, -0.00500367038572828, 0.00037602245962868136, 0.00797194141584138, 0.012475600582547486, 0.006475616673318048, 0.006899521841357152, 0.008204833992446462, -0.0002780493038396041, -0.009471656677002708, -0.00781721225939691, 0.025706250220537186 ], "state_rho": [ 0.0029325426245729127, 0.9265175064404806, 0.938806007305781, 0.9280826350053152, 0.9287765026092529, 0.9266663392384847, 0.9301863809426626, 0.9323704888423284, 0.9297124246756235, 0.9302101731300354, 0.9340693553288778, 0.9324037134647369, 0.9265123655398687, 0.9321505973736445, 0.927787164847056, 0.9323761165142059, 0.9224280416965485 ], "credit_rho": [ 0.02234963719577839, 0.7719902147849401, 0.8432890474796295, 0.8447659462690353, 0.877406562368075, 0.8958166440327963, 0.9122767845789591, 0.9254603485266367, 0.9236042300860087, 0.9263056516647339, 0.9323819329341253, 0.9314924577871958, 0.9266939163208008, 0.929329847296079, 0.9257104198137919, 0.9289217789967855, 0.9269137680530548 ], "dfa_nudge": [ -0.0003799900102118651, -0.0009698765352368355, -0.0013957968913018703, 0.00035563452790180844, -0.0003614289841304223, 0.0008372832089662552, -0.0013908503266672294, -0.0013163429684937, 0.0005854369762043158, -0.000560786963130037, 0.00015371766251822314, -1.0695696497956911e-05, 2.807355485856533e-05, 5.036763225992521e-05, 0.0008217894161740938, 0.0017250357971837123, -0.0004334271264572938 ], "state_nudge": [ -0.002327537008871635, -0.34193428109089535, -0.340603639682134, -0.35042013972997665, -0.3389856591820717, -0.34554000198841095, -0.3481511374314626, -0.3558509051799774, -0.3337947155038516, -0.33715616663297016, -0.33583804468313855, -0.35106155276298523, -0.34434546530246735, -0.34641525397698086, -0.3329972525437673, -0.34870391835769016, -0.3395152688026428 ], "credit_nudge": [ -0.0079942528779308, -0.30343081553777057, -0.31936119496822357, -0.33327333877484006, -0.3284662067890167, -0.3394550507267316, -0.3424832498033841, -0.3524467721581459, -0.33071904132763547, -0.3346964443723361, -0.3337800477941831, -0.34932391593853634, -0.34230031818151474, -0.3444634775320689, -0.3310972551504771, -0.34694332132736844, -0.3377286195755005 ], "bridge_residual": [], "state_bridge_loss": [ 66.01814270019531, 2.0012869834899902, 2.1027088165283203, 2.1019272804260254, 2.0727572441101074, 1.9770110845565796, 2.1761908531188965, 2.094480514526367, 1.9725749492645264, 2.0142102241516113, 2.0340821743011475, 1.9380583763122559, 1.989743947982788, 2.4057328701019287, 2.19437575340271, 2.1816155910491943, 2.0803794860839844 ], "credit_bridge_loss": [ 132.09298706054688, 11.307573318481445, 9.575517654418945, 8.768391609191895, 8.752981185913086, 9.271968841552734, 8.606213569641113, 8.753968238830566, 8.223488807678223, 9.935165405273438, 9.02967357635498, 8.346063613891602, 9.022041320800781, 8.36446762084961, 8.635725021362305, 8.760185241699219, 8.503408432006836 ], "term_loss": [ 111.63633728027344, 4.978545188903809, 3.8962953090667725, 3.81073260307312, 4.386394500732422, 4.507748603820801, 3.6740365028381348, 3.7450127601623535, 3.5060954093933105, 4.545898914337158, 4.322302341461182, 3.594371795654297, 4.038668632507324, 4.025404453277588, 3.9016177654266357, 3.7638583183288574, 3.807260036468506 ], "bridge_loss": [ 6.45359421014291e-07, 0.432157039642334, 0.20619139075279236, 0.24715952575206757, 0.1523856669664383, 0.12624874711036682, 0.13425695896148682, 0.16367560625076294, 0.14530189335346222, 0.18452416360378265, 0.11599670350551605, 0.11983858048915863, 0.11901542544364929, 0.16489851474761963, 0.16058529913425446, 0.09414370357990265, 0.12402483820915222 ], "term_grad_loss": [ 20.456655502319336, 5.8968706130981445, 5.4730305671691895, 4.710499286651611, 4.214200496673584, 4.637970924377441, 4.797920227050781, 4.845280170440674, 4.572091579437256, 5.204742908477783, 4.591374397277832, 4.631853103637695, 4.864356994628906, 4.174164772033691, 4.573522090911865, 4.9021830558776855, 4.5721235275268555 ], "fm_loss": [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ] }, "final_per_layer": { "dfa_costate_cos": [ -0.03974368795752525, -0.03234883397817612, -0.01280729565769434, -0.03282422572374344, 0.005050511099398136, -0.01393066719174385, 0.036346256732940674, 0.04377925395965576, 0.06324614584445953, 0.05324942618608475, -0.04297419637441635, -0.006917691323906183 ], "state_costate_cos": [ 0.9383341073989868, 0.9390445947647095, 0.9393846988677979, 0.9396730661392212, 0.9402315616607666, 0.940619707107544, 0.9413514137268066, 0.9420933723449707, 0.941848874092102, 0.9416857957839966, 0.9411274194717407, 0.9410437345504761 ], "credit_costate_cos": [ 0.937440037727356, 0.9379364252090454, 0.9384041428565979, 0.9385387897491455, 0.9392049312591553, 0.9398118257522583, 0.9403334259986877, 0.9412999153137207, 0.94156414270401, 0.9419326186180115, 0.9419490098953247, 0.9425133466720581 ], "dfa_rho": [ -0.023802103474736214, -0.039411380887031555, -0.008970402181148529, -0.0021191751584410667, 0.04573667049407959, 0.010564171709120274, 0.04995737224817276, 0.10094872862100601, 0.07801118493080139, 0.07188688963651657, -0.02006501331925392, 0.045738060027360916 ], "state_rho": [ 0.9204449653625488, 0.917449951171875, 0.9193954467773438, 0.9267550706863403, 0.9280773401260376, 0.9225738048553467, 0.9190236926078796, 0.924782931804657, 0.9248216152191162, 0.9217315912246704, 0.9215092658996582, 0.9225708246231079 ], "credit_rho": [ 0.9222818613052368, 0.923028826713562, 0.924055814743042, 0.9263180494308472, 0.9288915991783142, 0.922785758972168, 0.9231191873550415, 0.9254114627838135, 0.9321990013122559, 0.9311563372612, 0.934211015701294, 0.9295063018798828 ], "dfa_nudge": [ 0.014356113970279694, 0.013086749240756035, 0.004036703146994114, 0.011657552793622017, -0.0023863790556788445, 0.006091207265853882, -0.013204541988670826, -0.016102034598588943, -0.022989045828580856, -0.018843289464712143, 0.015521062538027763, 0.0035747764632105827 ], "state_nudge": [ -0.34063079953193665, -0.34022200107574463, -0.33977580070495605, -0.3401387929916382, -0.34019535779953003, -0.33964985609054565, -0.3392874002456665, -0.33898109197616577, -0.3392573297023773, -0.3390581011772156, -0.33809930086135864, -0.33888739347457886 ], "credit_nudge": [ -0.3375471234321594, -0.33736705780029297, -0.33710652589797974, -0.33756691217422485, -0.3378610908985138, -0.33758991956710815, -0.33741605281829834, -0.3374325633049011, -0.33806926012039185, -0.33822208642959595, -0.33765909075737, -0.3389057517051697 ] } }