{ "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": 456, "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": [ -0.008305357536301017, -0.0011448257913192113, -0.011490714969113469, -0.005118173896335065, -0.0028971168600643673, -0.007146042305976152, -0.005135039333254099, -0.006408803591815134, 0.0030692683843274913 ], "state_costate_cos": [ 0.010766413528472185, 0.9454400340716044, 0.9422353406747183, 0.945327232281367, 0.9411078443129858, 0.9500264773766199, 0.947187085946401, 0.9483370830615362, 0.9451590329408646 ], "credit_costate_cos": [ 0.010942678588132063, 0.8771380881468455, 0.9226242254177729, 0.9362819741169611, 0.9372822741667429, 0.9464249561230341, 0.9468309829632441, 0.9452949613332748, 0.9439582029978434 ], "dfa_rho": [ -0.0028248391657446823, -0.001316564545656244, -0.00823917348558704, -0.008288809360237792, -0.009200356512640914, -0.02464000484906137, 0.006656331475824118, -0.014378171879798174, 0.00816792449525868 ], "state_rho": [ 0.02655455912463367, 0.9340365380048752, 0.930340642730395, 0.9328931520382563, 0.9261527856191, 0.9353549828131994, 0.9365303913752238, 0.930912658572197, 0.9325708796580633 ], "credit_rho": [ 0.015292729716748, 0.8341556191444397, 0.884218767285347, 0.9175956894954046, 0.9214809189240137, 0.9294924139976501, 0.9364803830782572, 0.9272431135177612, 0.9332626809676489 ], "dfa_nudge": [ 0.004384364855165283, 0.0014555706487347682, 0.005660574107120435, 0.003231095770994822, 0.0024609332904219627, 0.0033611954810718694, 0.0025722047624488673, 0.003350513521581888, -0.00046457063096265 ], "state_nudge": [ -0.005116054400180777, -0.36389906456073123, -0.37967536846796673, -0.3594902977347374, -0.3901909242073695, -0.3562925284107526, -0.34591514120499295, -0.3550695503751437, -0.3515613650282224 ], "credit_nudge": [ -0.003232262640570601, -0.34101804345846176, -0.37139702836672467, -0.355108546713988, -0.38709117472171783, -0.3536044582724571, -0.34420712540547055, -0.35209985077381134, -0.3494671831528346 ], "bridge_residual": [], "state_bridge_loss": [ 66.31975555419922, 2.3091068267822266, 2.0314407348632812, 2.110802412033081, 1.9804980754852295, 1.8128150701522827, 2.0147881507873535, 2.169416904449463, 2.0117833614349365 ], "credit_bridge_loss": [ 158.73072814941406, 11.18570613861084, 9.38658332824707, 9.429727554321289, 10.842954635620117, 10.344818115234375, 10.250753402709961, 10.136574745178223, 9.871820449829102 ], "term_loss": [ 132.93673706054688, 4.870186805725098, 3.9316887855529785, 4.335302829742432, 5.871437072753906, 4.5994768142700195, 5.077899932861328, 4.8271284103393555, 4.638625621795654 ], "bridge_loss": [ 7.166463547036983e-07, 0.3530547022819519, 0.20790576934814453, 0.15763649344444275, 0.18163490295410156, 0.154127299785614, 0.1578863561153412, 0.13594242930412292, 0.12283627688884735 ], "term_grad_loss": [ 25.793991088867188, 5.962464809417725, 5.246988296508789, 4.936788082122803, 4.789882183074951, 5.591214656829834, 5.0149664878845215, 5.173503875732422, 5.110358238220215 ], "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": [ 0.0057477825321257114, -0.05385865271091461, 0.031865090131759644, -0.0450170561671257, 0.01222146674990654, 0.048112209886312485, -0.0029807062819600105, 0.02491983398795128, -0.03023526817560196, -0.0383906215429306, 0.005803969223052263, 0.07864317297935486 ], "state_costate_cos": [ 0.9431054592132568, 0.9437580108642578, 0.9442368149757385, 0.9451977014541626, 0.9453786611557007, 0.9453213214874268, 0.9453005790710449, 0.945610761642456, 0.9459618330001831, 0.946029007434845, 0.9460264444351196, 0.9459818005561829 ], "credit_costate_cos": [ 0.9404189586639404, 0.9415367841720581, 0.9419513940811157, 0.9428710341453552, 0.943200409412384, 0.9437626004219055, 0.9443795680999756, 0.9448325037956238, 0.945106029510498, 0.9459753036499023, 0.9465078115463257, 0.9469560384750366 ], "dfa_rho": [ 0.04672882705926895, -0.030157513916492462, 0.03139471262693405, -0.02426629513502121, -0.003147948533296585, 0.06672847270965576, -0.0028628872241824865, -0.0006099608726799488, -0.06767985224723816, -0.07952223718166351, 0.023338939994573593, 0.13807083666324615 ], "state_rho": [ 0.9322999715805054, 0.9332318902015686, 0.933842658996582, 0.9367775917053223, 0.9355518221855164, 0.9306066036224365, 0.9324768781661987, 0.9316605925559998, 0.9233090281486511, 0.9335967302322388, 0.9341922998428345, 0.933304488658905 ], "credit_rho": [ 0.929013192653656, 0.925915002822876, 0.9249007701873779, 0.9293269515037537, 0.9353340268135071, 0.9355732202529907, 0.9336704611778259, 0.9395359754562378, 0.934350848197937, 0.9350711107254028, 0.9423503279685974, 0.9341102838516235 ], "dfa_nudge": [ -0.0025230227038264275, 0.01772259920835495, -0.011610012501478195, 0.01727830246090889, -0.002929902635514736, -0.01679525338113308, 0.0033558662980794907, -0.007272847928106785, 0.013050400651991367, 0.014434966258704662, -0.0013312064111232758, -0.02895473688840866 ], "state_nudge": [ -0.3518102467060089, -0.3520665764808655, -0.35131847858428955, -0.35165226459503174, -0.35131698846817017, -0.35155874490737915, -0.3516783118247986, -0.3522875905036926, -0.35148167610168457, -0.3511171340942383, -0.35132837295532227, -0.3511199951171875 ], "credit_nudge": [ -0.34800052642822266, -0.3485172390937805, -0.3479737639427185, -0.3486989140510559, -0.3486787676811218, -0.3492557406425476, -0.3497552275657654, -0.35073035955429077, -0.35016047954559326, -0.3501723110675812, -0.3507159650325775, -0.35094690322875977 ] } }