{ "config": { "d_hidden": 64, "output_dim": 10, "num_layers": 12, "sigma": 0.03, "batch_size": 256, "num_steps": 5000, "lr_fb": 0.001, "lam": 1.0, "K": 8, "ema_momentum": 0.995, "sigma_bridge": 0.3, "eval_every": 1000, "seed": 42, "gpu": 0, "output_dir": "results/toy_lq", "vnet_hidden": 256, "vnet_layers": 3, "term_grad_weight": 0.0, "fm_weight": 0.0 }, "log": { "steps": [ 1, 1000, 2000, 3000, 4000, 5000 ], "dfa_costate_cos": [ 0.001022340264171362, 0.0024294707691296935, 0.000613357910575966, 0.002641987521201372, 0.0019003628791930776, 0.004648004221962765 ], "state_costate_cos": [ 0.009337988216429949, 0.9437081764141718, 0.9386202742656072, 0.9438605606555939, 0.9475679496924082, 0.9496726940075556 ], "credit_costate_cos": [ 0.023481125943362713, 0.0021192015459140143, 0.05908452393487096, 0.11570050132771333, 0.09644180163741112, 0.06304322431484859 ], "dfa_rho": [ 0.015879416760678094, 0.009290086299491426, 0.0009949249991526206, -0.004670841522359599, -0.0029721508423487344, 0.0010571565168599288 ], "state_rho": [ 0.0029325426245729127, 0.9371241927146912, 0.9219773809115092, 0.9298903445402781, 0.9345368842283884, 0.9332165767749151 ], "credit_rho": [ 0.02086908878603329, -0.014479975526531538, 0.04267269264285763, 0.10674913817395766, 0.091526560485363, 0.04765695089008659 ], "dfa_nudge": [ -0.0003799900102118651, -0.0008909914953013262, 0.00031573620314399403, -0.0008827850688248873, -0.0003006396194299062, -0.0016971436173965533 ], "state_nudge": [ -0.002327537008871635, -0.33619146794080734, -0.3439306889971097, -0.32351043323675793, -0.33487510432799655, -0.35304194688796997 ], "credit_nudge": [ -0.007470574385176103, 0.002750888311614593, -0.017584003585701186, -0.037241545505821705, -0.03454847944279512, -0.02138534157226483 ], "bridge_residual": [], "state_bridge_loss": [ 66.01814270019531, 2.143366813659668, 1.9674744606018066, 2.152421712875366, 2.0728020668029785, 2.0048255920410156 ], "credit_bridge_loss": [ 111.63633728027344, 0.35802197456359863, 0.10368431359529495, 0.12013768404722214, 0.07032017409801483, 0.051539346575737 ], "term_loss": [ 111.63633728027344, 0.2640027403831482, 0.05911973863840103, 0.06323631852865219, 0.03984691575169563, 0.03099522553384304 ], "bridge_loss": [ 3.0909704946679994e-06, 0.09401924908161163, 0.04456457495689392, 0.056901365518569946, 0.030473260208964348, 0.02054412104189396 ], "term_grad_loss": [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ], "fm_loss": [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ] }, "final_per_layer": { "dfa_costate_cos": [ -0.025100916624069214, -0.025588383898139, 0.0002230944810435176, -0.012649456970393658, -0.003438096959143877, -0.008626529946923256, 0.04746557027101517, 0.0495537593960762, 0.05357586592435837, 0.039518602192401886, -0.06593793630599976, 0.0067804791033267975 ], "state_costate_cos": [ 0.94776451587677, 0.9486032724380493, 0.9489368200302124, 0.9489387273788452, 0.9493762850761414, 0.9498711228370667, 0.9505667686462402, 0.9509860277175903, 0.9503088593482971, 0.9503939151763916, 0.9501732587814331, 0.9501527547836304 ], "credit_costate_cos": [ 0.06872019171714783, 0.0653301253914833, 0.06267654895782471, 0.060872238129377365, 0.059321627020835876, 0.05934217572212219, 0.06014599651098251, 0.05938819795846939, 0.06065107509493828, 0.06321083009243011, 0.07017489522695541, 0.066684789955616 ], "dfa_rho": [ -0.039396628737449646, -0.062005721032619476, 0.005523890256881714, 0.014756813645362854, -0.029910940676927567, 0.016373179852962494, 0.06027424708008766, 0.07274787873029709, 0.04506715014576912, 0.02043943479657173, -0.07323547452688217, -0.017947951331734657 ], "state_rho": [ 0.9350653886795044, 0.9352835416793823, 0.9357374310493469, 0.934054434299469, 0.9226828813552856, 0.930233359336853, 0.9396862983703613, 0.9324671030044556, 0.9295884966850281, 0.9347179532051086, 0.9344969987869263, 0.9345850348472595 ], "credit_rho": [ 0.07084883749485016, 0.0004139472730457783, 0.08371011912822723, 0.010933063924312592, 0.07074157148599625, 0.02977888286113739, 0.06011004000902176, 0.020937703549861908, 0.067134790122509, 0.04539356008172035, 0.06114630401134491, 0.050734590739011765 ], "dfa_nudge": [ 0.010068551637232304, 0.009215200319886208, -0.002217007800936699, 0.007142472080886364, 0.0020765019580721855, 0.004165132530033588, -0.019989464432001114, -0.018945707008242607, -0.01947549544274807, -0.014030318707227707, 0.023684613406658173, -0.0020602019503712654 ], "state_nudge": [ -0.3548189401626587, -0.3539973497390747, -0.3535424768924713, -0.35353928804397583, -0.35366636514663696, -0.35295385122299194, -0.35282424092292786, -0.35278239846229553, -0.3528212308883667, -0.3526480793952942, -0.35135188698768616, -0.35155725479125977 ], "credit_nudge": [ -0.024559948593378067, -0.023154649883508682, -0.02181980386376381, -0.020881079137325287, -0.02010848931968212, -0.019884146749973297, -0.020017635077238083, -0.01966693066060543, -0.020179908722639084, -0.02114756405353546, -0.023488853126764297, -0.021715089678764343 ] } }