iamshlomo commited on
Commit
b51fd6e
·
verified ·
1 Parent(s): b2884f8

Upload generalization/full_analysis_20260324_1116/minihack_full_results.json with huggingface_hub

Browse files
generalization/full_analysis_20260324_1116/minihack_full_results.json ADDED
@@ -0,0 +1,2005 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "run_id": "20260324_1116",
3
+ "max_iter": 500,
4
+ "eval_eps": 10,
5
+ "id_envs": [
6
+ "MiniHack-Room-Random-5x5-v0",
7
+ "MiniHack-Room-Random-15x15-v0",
8
+ "MiniHack-Corridor-R2-v0",
9
+ "MiniHack-MazeWalk-9x9-v0"
10
+ ],
11
+ "ood_envs": [
12
+ "MiniHack-Room-Dark-15x15-v0",
13
+ "MiniHack-Corridor-R5-v0",
14
+ "MiniHack-MazeWalk-45x19-v0"
15
+ ],
16
+ "baseline_id": 0.45,
17
+ "baseline_ood": 0.03333333333333333,
18
+ "baseline_rl_id": 0.025,
19
+ "baseline_rl_ood": 0.03333333333333333,
20
+ "kl_penalty": {
21
+ "id": 0.425,
22
+ "ood": 0.10000000000000002
23
+ },
24
+ "frozen_backbone": {
25
+ "id": 0.125,
26
+ "ood": 0.0
27
+ },
28
+ "bc_on_wins": {
29
+ "id": 0.125,
30
+ "ood": 0.03333333333333333
31
+ },
32
+ "low_t_only": {
33
+ "id": 0.05,
34
+ "ood": 0.0
35
+ },
36
+ "synthetic": {
37
+ "id": 0.05
38
+ },
39
+ "histories": {
40
+ "baseline_rl": {
41
+ "iter": [
42
+ 10,
43
+ 20,
44
+ 30,
45
+ 40,
46
+ 50,
47
+ 60,
48
+ 70,
49
+ 80,
50
+ 90,
51
+ 100,
52
+ 110,
53
+ 120,
54
+ 130,
55
+ 140,
56
+ 150,
57
+ 160,
58
+ 170,
59
+ 180,
60
+ 190,
61
+ 200,
62
+ 210,
63
+ 220,
64
+ 230,
65
+ 240,
66
+ 250,
67
+ 260,
68
+ 270,
69
+ 280,
70
+ 290,
71
+ 300,
72
+ 310,
73
+ 320,
74
+ 330,
75
+ 340,
76
+ 350,
77
+ 360,
78
+ 370,
79
+ 380,
80
+ 390,
81
+ 400,
82
+ 410,
83
+ 420,
84
+ 430,
85
+ 440,
86
+ 450,
87
+ 460,
88
+ 470,
89
+ 480,
90
+ 490,
91
+ 500
92
+ ],
93
+ "loss": [
94
+ 0.3538270264863968,
95
+ 0.4377899743616581,
96
+ 0.08652612939476967,
97
+ 0.04114371752366423,
98
+ 0.03957269466482103,
99
+ 0.046368783246725796,
100
+ 0.029016940807923674,
101
+ 0.023031661449931563,
102
+ 0.018955788156017662,
103
+ 0.012367414694745093,
104
+ 0.038063708273693916,
105
+ 0.02385816101450473,
106
+ 0.0201659549260512,
107
+ 0.017254522698931397,
108
+ 0.010255878919269889,
109
+ 0.008153129415586591,
110
+ 0.008980516204610467,
111
+ 0.0018633072701049968,
112
+ 0.005288890225347132,
113
+ 0.0032623883977066725,
114
+ 0.007189091545296833,
115
+ 0.0036865342379314827,
116
+ 0.034503579896409065,
117
+ 0.0024017324147280307,
118
+ 0.008079327135055792,
119
+ 0.005166706864838488,
120
+ 0.004526812941185199,
121
+ 0.005694168864283711,
122
+ 0.005634190852288157,
123
+ 0.025248501528403723,
124
+ 0.014854160777758807,
125
+ 0.011417066992726177,
126
+ 0.02595224802644225,
127
+ 0.0027896528499695705,
128
+ 0.014399234169104602,
129
+ 0.006570604163061944,
130
+ 0.011021597406215732,
131
+ 0.003368453289658646,
132
+ 0.007000996673014015,
133
+ 0.0020966271125871573,
134
+ 0.0012383776893329923,
135
+ 0.0010686448353226297,
136
+ 0.0005573026563070016,
137
+ 0.0013555235869716852,
138
+ 0.003491711810420384,
139
+ 0.0008705336511411587,
140
+ 0.003108983759284456,
141
+ 0.000381374674361723,
142
+ 0.0027816690466352157,
143
+ 0.0010556293743320567
144
+ ],
145
+ "mean_return": [
146
+ 2.261000000000013,
147
+ -4.155999999999993,
148
+ -4.689999999999982,
149
+ -3.7719999999999976,
150
+ -1.9609999999999892,
151
+ -3.9939999999999976,
152
+ -3.5319999999999903,
153
+ 1.952000000000006,
154
+ 1.8880000000000041,
155
+ -3.4559999999999973,
156
+ 5.916000000000009,
157
+ -1.094999999999996,
158
+ -4.045999999999998,
159
+ -1.710999999999991,
160
+ -0.394999999999999,
161
+ -1.3489999999999935,
162
+ -1.7289999999999899,
163
+ 0.892000000000005,
164
+ -4.261999999999992,
165
+ -3.395999999999995,
166
+ -4.481999999999983,
167
+ -2.3579999999999943,
168
+ -0.8409999999999996,
169
+ -4.22599999999999,
170
+ -3.3879999999999897,
171
+ -3.4499999999999993,
172
+ -0.9589999999999945,
173
+ -2.781999999999993,
174
+ -3.671999999999991,
175
+ 0.8160000000000072,
176
+ -0.30299999999999183,
177
+ -3.2180000000000013,
178
+ -3.4339999999999895,
179
+ -2.883999999999995,
180
+ -4.223999999999992,
181
+ -1.0749999999999986,
182
+ -4.053999999999989,
183
+ -3.3539999999999885,
184
+ -3.3560000000000003,
185
+ -1.3789999999999962,
186
+ -3.493999999999994,
187
+ 0.8280000000000047,
188
+ -4.401999999999983,
189
+ 1.2080000000000062,
190
+ 1.5699999999999998,
191
+ -4.139999999999992,
192
+ -4.19799999999999,
193
+ -4.13599999999999,
194
+ -3.729999999999994,
195
+ -0.9930000000000012
196
+ ],
197
+ "win_rate_episode": [
198
+ 0.2,
199
+ 0.0,
200
+ 0.0,
201
+ 0.0,
202
+ 0.1,
203
+ 0.0,
204
+ 0.0,
205
+ 0.2,
206
+ 0.2,
207
+ 0.0,
208
+ 0.4,
209
+ 0.1,
210
+ 0.0,
211
+ 0.1,
212
+ 0.1,
213
+ 0.1,
214
+ 0.1,
215
+ 0.2,
216
+ 0.0,
217
+ 0.0,
218
+ 0.0,
219
+ 0.0,
220
+ 0.1,
221
+ 0.0,
222
+ 0.0,
223
+ 0.0,
224
+ 0.1,
225
+ 0.0,
226
+ 0.0,
227
+ 0.2,
228
+ 0.1,
229
+ 0.0,
230
+ 0.0,
231
+ 0.0,
232
+ 0.0,
233
+ 0.1,
234
+ 0.0,
235
+ 0.0,
236
+ 0.0,
237
+ 0.1,
238
+ 0.0,
239
+ 0.2,
240
+ 0.0,
241
+ 0.2,
242
+ 0.2,
243
+ 0.0,
244
+ 0.0,
245
+ 0.0,
246
+ 0.0,
247
+ 0.1
248
+ ],
249
+ "id_winrate": [
250
+ 0.07500000000000001,
251
+ 0.05,
252
+ 0.025,
253
+ 0.075,
254
+ 0.025
255
+ ],
256
+ "id_winrate_iter": [
257
+ 100,
258
+ 200,
259
+ 300,
260
+ 400,
261
+ 500
262
+ ],
263
+ "ood_winrate": [
264
+ 0.0
265
+ ],
266
+ "ood_winrate_iter": [
267
+ 500
268
+ ],
269
+ "grad_align_iter": [],
270
+ "grad_align": [],
271
+ "rl_grad_norm": [],
272
+ "bc_grad_norm": [],
273
+ "repr_drift_iter": [
274
+ 25,
275
+ 50,
276
+ 75,
277
+ 100,
278
+ 125,
279
+ 150,
280
+ 175,
281
+ 200,
282
+ 225,
283
+ 250,
284
+ 275,
285
+ 300,
286
+ 325,
287
+ 350,
288
+ 375,
289
+ 400,
290
+ 425,
291
+ 450,
292
+ 475,
293
+ 500
294
+ ],
295
+ "repr_drift": [
296
+ 10.160261154174805,
297
+ 7.573764324188232,
298
+ 9.211956024169922,
299
+ 11.56298828125,
300
+ 9.587845802307129,
301
+ 10.252934455871582,
302
+ 9.261374473571777,
303
+ 8.68515396118164,
304
+ 10.217469215393066,
305
+ 9.4309663772583,
306
+ 9.95379638671875,
307
+ 14.22205638885498,
308
+ 14.815824508666992,
309
+ 16.02154541015625,
310
+ 11.853056907653809,
311
+ 13.566880226135254,
312
+ 13.512826919555664,
313
+ 12.569154739379883,
314
+ 14.0609769821167,
315
+ 11.774737358093262
316
+ ],
317
+ "t_analysis_iter": [
318
+ 25,
319
+ 50,
320
+ 75,
321
+ 100,
322
+ 125,
323
+ 150,
324
+ 175,
325
+ 200,
326
+ 225,
327
+ 250,
328
+ 275,
329
+ 300,
330
+ 325,
331
+ 350,
332
+ 375,
333
+ 400,
334
+ 425,
335
+ 450,
336
+ 475,
337
+ 500
338
+ ],
339
+ "norm_low_t": [
340
+ 0.5284388065338135,
341
+ 0.2081630975008011,
342
+ 0.17324304580688477,
343
+ 0.017162062227725983,
344
+ 0.0803459957242012,
345
+ 0.008837013505399227,
346
+ 0.2382400929927826,
347
+ 0.10077612847089767,
348
+ 0.17042259871959686,
349
+ 0.17041170597076416,
350
+ 0.032309841364622116,
351
+ 0.38514673709869385,
352
+ 0.05558478832244873,
353
+ 0.00013036349264439195,
354
+ 0.0001973858888959512,
355
+ 0.00022925512166693807,
356
+ 6.406019383575767e-05,
357
+ 0.24417723715305328,
358
+ 0.0019458290189504623,
359
+ 0.002486016834154725
360
+ ],
361
+ "norm_high_t": [
362
+ 0.5880557894706726,
363
+ 0.2764274477958679,
364
+ 0.15763837099075317,
365
+ 0.19735455513000488,
366
+ 0.1485709697008133,
367
+ 0.16311271488666534,
368
+ 0.10455828905105591,
369
+ 0.09554796665906906,
370
+ 0.11885329335927963,
371
+ 0.19188638031482697,
372
+ 0.15504693984985352,
373
+ 0.3395625948905945,
374
+ 0.06264150887727737,
375
+ 2.0697410106658936,
376
+ 0.18694882094860077,
377
+ 0.003971199970692396,
378
+ 0.0015933484537526965,
379
+ 0.1253371685743332,
380
+ 0.008088835515081882,
381
+ 0.03807271271944046
382
+ ],
383
+ "lowhigh_cos": [
384
+ 0.662343442440033,
385
+ 0.5513224601745605,
386
+ -0.13289889693260193,
387
+ 0.525389552116394,
388
+ 0.5405603051185608,
389
+ -0.36270010471343994,
390
+ 0.24438165128231049,
391
+ -0.5304811000823975,
392
+ 0.7923681735992432,
393
+ 0.1763698160648346,
394
+ 0.41906774044036865,
395
+ 0.47982245683670044,
396
+ 0.5349423885345459,
397
+ -0.20126745104789734,
398
+ -0.012673936784267426,
399
+ 0.3375846743583679,
400
+ 0.6472594738006592,
401
+ 0.7772222757339478,
402
+ 0.3886796832084656,
403
+ 0.6496570110321045
404
+ ]
405
+ },
406
+ "kl_penalty": {
407
+ "iter": [
408
+ 10,
409
+ 20,
410
+ 30,
411
+ 40,
412
+ 50,
413
+ 60,
414
+ 70,
415
+ 80,
416
+ 90,
417
+ 100,
418
+ 110,
419
+ 120,
420
+ 130,
421
+ 140,
422
+ 150,
423
+ 160,
424
+ 170,
425
+ 180,
426
+ 190,
427
+ 200,
428
+ 210,
429
+ 220,
430
+ 230,
431
+ 240,
432
+ 250,
433
+ 260,
434
+ 270,
435
+ 280,
436
+ 290,
437
+ 300,
438
+ 310,
439
+ 320,
440
+ 330,
441
+ 340,
442
+ 350,
443
+ 360,
444
+ 370,
445
+ 380,
446
+ 390,
447
+ 400,
448
+ 410,
449
+ 420,
450
+ 430,
451
+ 440,
452
+ 450,
453
+ 460,
454
+ 470,
455
+ 480,
456
+ 490,
457
+ 500
458
+ ],
459
+ "loss": [
460
+ 0.7557014524936676,
461
+ 0.3678558230400085,
462
+ 0.2518810033798218,
463
+ 0.35838921666145324,
464
+ 0.1938560798764229,
465
+ 0.23091877549886702,
466
+ 0.2477170094847679,
467
+ 0.19129770025610923,
468
+ 0.26212684661149976,
469
+ 0.20293212532997132,
470
+ 0.202893528342247,
471
+ 0.2635331779718399,
472
+ 0.19642161577939987,
473
+ 0.3245967239141464,
474
+ 0.2945104017853737,
475
+ 0.23478922098875046,
476
+ 0.23234525918960572,
477
+ 0.3016815945506096,
478
+ 0.32549490183591845,
479
+ 0.2518655315041542,
480
+ 0.2638390526175499,
481
+ 0.243428336083889,
482
+ 0.29855785220861436,
483
+ 0.2755287617444992,
484
+ 0.38007437586784365,
485
+ 0.28243361264467237,
486
+ 0.32891527116298674,
487
+ 0.36257472485303877,
488
+ 0.4055925294756889,
489
+ 0.243349352478981,
490
+ 0.4049526423215866,
491
+ 0.27672329396009443,
492
+ 0.3030186653137207,
493
+ 0.26211272180080414,
494
+ 0.28421989679336546,
495
+ 0.33425604552030563,
496
+ 0.39837211817502977,
497
+ 0.3984354332089424,
498
+ 0.3670164056122303,
499
+ 0.24117564111948014,
500
+ 0.3429794669151306,
501
+ 0.4218721553683281,
502
+ 0.4734982058405876,
503
+ 0.3788636654615402,
504
+ 0.3556398391723633,
505
+ 0.3187734082341194,
506
+ 0.3146557629108429,
507
+ 0.3174664542078972,
508
+ 0.27082621604204177,
509
+ 0.32873593270778656
510
+ ],
511
+ "mean_return": [
512
+ 7.125000000000003,
513
+ 15.425,
514
+ 13.065000000000001,
515
+ 9.499000000000006,
516
+ 11.684000000000008,
517
+ 11.327000000000005,
518
+ 11.770000000000007,
519
+ 11.629000000000005,
520
+ 9.21000000000001,
521
+ 10.411999999999999,
522
+ 12.987000000000005,
523
+ 14.014999999999997,
524
+ 10.274,
525
+ 4.512000000000006,
526
+ 7.4760000000000035,
527
+ 9.329000000000004,
528
+ 3.5090000000000154,
529
+ 4.014000000000007,
530
+ 7.037000000000004,
531
+ 6.2900000000000045,
532
+ 4.163000000000007,
533
+ 7.446999999999998,
534
+ 7.409999999999999,
535
+ -0.7689999999999918,
536
+ 3.015000000000003,
537
+ 10.074000000000003,
538
+ 2.8990000000000054,
539
+ 12.178000000000003,
540
+ 0.8730000000000111,
541
+ 2.228000000000012,
542
+ 4.342000000000007,
543
+ 5.358,
544
+ 7.877000000000001,
545
+ 8.392000000000014,
546
+ 8.838000000000006,
547
+ 3.274000000000008,
548
+ 6.922000000000006,
549
+ 2.493000000000001,
550
+ 9.593,
551
+ 9.479000000000006,
552
+ 0.4620000000000126,
553
+ 7.5969999999999995,
554
+ 15.603,
555
+ 0.6589999999999905,
556
+ 8.977000000000007,
557
+ 12.063999999999998,
558
+ 10.15600000000001,
559
+ 1.462000000000001,
560
+ 3.137000000000004,
561
+ 9.123000000000006
562
+ ],
563
+ "win_rate_episode": [
564
+ 0.4,
565
+ 0.7,
566
+ 0.6,
567
+ 0.5,
568
+ 0.6,
569
+ 0.6,
570
+ 0.6,
571
+ 0.6,
572
+ 0.5,
573
+ 0.5,
574
+ 0.6,
575
+ 0.6,
576
+ 0.5,
577
+ 0.3,
578
+ 0.4,
579
+ 0.5,
580
+ 0.3,
581
+ 0.3,
582
+ 0.4,
583
+ 0.3,
584
+ 0.3,
585
+ 0.4,
586
+ 0.4,
587
+ 0.1,
588
+ 0.2,
589
+ 0.6,
590
+ 0.2,
591
+ 0.6,
592
+ 0.2,
593
+ 0.2,
594
+ 0.3,
595
+ 0.3,
596
+ 0.4,
597
+ 0.5,
598
+ 0.5,
599
+ 0.2,
600
+ 0.4,
601
+ 0.2,
602
+ 0.5,
603
+ 0.5,
604
+ 0.2,
605
+ 0.4,
606
+ 0.7,
607
+ 0.2,
608
+ 0.5,
609
+ 0.6,
610
+ 0.5,
611
+ 0.2,
612
+ 0.2,
613
+ 0.5
614
+ ],
615
+ "id_winrate": [
616
+ 0.7000000000000001,
617
+ 0.275,
618
+ 0.44999999999999996,
619
+ 0.4,
620
+ 0.525
621
+ ],
622
+ "id_winrate_iter": [
623
+ 100,
624
+ 200,
625
+ 300,
626
+ 400,
627
+ 500
628
+ ],
629
+ "ood_winrate": [
630
+ 0.03333333333333333
631
+ ],
632
+ "ood_winrate_iter": [
633
+ 500
634
+ ],
635
+ "grad_align_iter": [],
636
+ "grad_align": [],
637
+ "rl_grad_norm": [],
638
+ "bc_grad_norm": [],
639
+ "repr_drift_iter": [
640
+ 25,
641
+ 50,
642
+ 75,
643
+ 100,
644
+ 125,
645
+ 150,
646
+ 175,
647
+ 200,
648
+ 225,
649
+ 250,
650
+ 275,
651
+ 300,
652
+ 325,
653
+ 350,
654
+ 375,
655
+ 400,
656
+ 425,
657
+ 450,
658
+ 475,
659
+ 500
660
+ ],
661
+ "repr_drift": [
662
+ 0.36094218492507935,
663
+ 0.5449057221412659,
664
+ 0.4667113423347473,
665
+ 0.34563779830932617,
666
+ 0.37591034173965454,
667
+ 0.35121041536331177,
668
+ 0.3511926829814911,
669
+ 0.5004967451095581,
670
+ 0.3912479877471924,
671
+ 0.8314864039421082,
672
+ 0.6106400489807129,
673
+ 0.7350732088088989,
674
+ 0.5971804857254028,
675
+ 0.3814377784729004,
676
+ 0.5749532580375671,
677
+ 0.5550214052200317,
678
+ 0.5665072798728943,
679
+ 0.766390323638916,
680
+ 0.8498607277870178,
681
+ 0.7186318039894104
682
+ ],
683
+ "t_analysis_iter": [
684
+ 25,
685
+ 50,
686
+ 75,
687
+ 100,
688
+ 125,
689
+ 150,
690
+ 175,
691
+ 200,
692
+ 225,
693
+ 250,
694
+ 275,
695
+ 300,
696
+ 325,
697
+ 350,
698
+ 375,
699
+ 400,
700
+ 425,
701
+ 450,
702
+ 475,
703
+ 500
704
+ ],
705
+ "norm_low_t": [
706
+ 1.0817543268203735,
707
+ 0.7415452003479004,
708
+ 0.31479886174201965,
709
+ 0.40329068899154663,
710
+ 0.7726807594299316,
711
+ 0.8081589937210083,
712
+ 4.519750118255615,
713
+ 0.3104645609855652,
714
+ 0.35659709572792053,
715
+ 5.784140586853027,
716
+ 5.175623416900635,
717
+ 0.3377837836742401,
718
+ 0.35866743326187134,
719
+ 1.6006625890731812,
720
+ 0.39933130145072937,
721
+ 1.962070345878601,
722
+ 0.5190642476081848,
723
+ 4.11475133895874,
724
+ 0.9334627985954285,
725
+ 0.3140447735786438
726
+ ],
727
+ "norm_high_t": [
728
+ 1.1227264404296875,
729
+ 0.8195191025733948,
730
+ 3.2485482692718506,
731
+ 0.7025418877601624,
732
+ 5.411553382873535,
733
+ 4.700546741485596,
734
+ 2.057323932647705,
735
+ 0.5917654037475586,
736
+ 3.6300909519195557,
737
+ 2.854146957397461,
738
+ 1.478105068206787,
739
+ 1.4446972608566284,
740
+ 0.6589544415473938,
741
+ 3.9933993816375732,
742
+ 0.5964608192443848,
743
+ 1.9211534261703491,
744
+ 0.6782035827636719,
745
+ 2.1749870777130127,
746
+ 4.211155414581299,
747
+ 0.5032972097396851
748
+ ],
749
+ "lowhigh_cos": [
750
+ 0.6076641082763672,
751
+ 0.4695075750350952,
752
+ 0.2275436371564865,
753
+ 0.2731117606163025,
754
+ 0.1992029994726181,
755
+ 0.12773582339286804,
756
+ 0.06728317588567734,
757
+ 0.3145290017127991,
758
+ 0.13208569586277008,
759
+ 0.5987216234207153,
760
+ 0.44497546553611755,
761
+ -0.15342308580875397,
762
+ 0.1819535493850708,
763
+ -0.3881320655345917,
764
+ 0.5761399269104004,
765
+ 0.5740898847579956,
766
+ 0.5706593990325928,
767
+ -0.17910754680633545,
768
+ 0.3718031942844391,
769
+ -0.12967117130756378
770
+ ]
771
+ },
772
+ "frozen_backbone": {
773
+ "iter": [
774
+ 10,
775
+ 20,
776
+ 30,
777
+ 40,
778
+ 50,
779
+ 60,
780
+ 70,
781
+ 80,
782
+ 90,
783
+ 100,
784
+ 110,
785
+ 120,
786
+ 130,
787
+ 140,
788
+ 150,
789
+ 160,
790
+ 170,
791
+ 180,
792
+ 190,
793
+ 200,
794
+ 210,
795
+ 220,
796
+ 230,
797
+ 240,
798
+ 250,
799
+ 260,
800
+ 270,
801
+ 280,
802
+ 290,
803
+ 300,
804
+ 310,
805
+ 320,
806
+ 330,
807
+ 340,
808
+ 350,
809
+ 360,
810
+ 370,
811
+ 380,
812
+ 390,
813
+ 400,
814
+ 410,
815
+ 420,
816
+ 430,
817
+ 440,
818
+ 450,
819
+ 460,
820
+ 470,
821
+ 480,
822
+ 490,
823
+ 500
824
+ ],
825
+ "loss": [
826
+ 0.6142368137836456,
827
+ 0.7526629269123077,
828
+ 0.5536824703216553,
829
+ 0.8061215162277222,
830
+ 0.6501811683177948,
831
+ 0.9902062594890595,
832
+ 0.8726756662130356,
833
+ 0.8145544409751893,
834
+ 0.7573557883501053,
835
+ 1.2141560435295105,
836
+ 0.5211864888668061,
837
+ 0.5383211493492126,
838
+ 0.8611633062362671,
839
+ 0.6769385233521461,
840
+ 0.8571627408266067,
841
+ 0.7400204911828041,
842
+ 0.35199219882488253,
843
+ 0.5443531692028045,
844
+ 0.48275753408670424,
845
+ 0.4545718625187874,
846
+ 0.5590669482946395,
847
+ 0.2528196260333061,
848
+ 0.2803684830665588,
849
+ 0.2787619173526764,
850
+ 0.3422154411673546,
851
+ 0.35092500075697897,
852
+ 0.29189412444829943,
853
+ 0.3062771186232567,
854
+ 0.1682231456041336,
855
+ 0.4048779964447021,
856
+ 0.15562354549765586,
857
+ 0.15582111850380898,
858
+ 0.27918287068605424,
859
+ 0.2198553279042244,
860
+ 0.22693262323737146,
861
+ 0.16309556886553764,
862
+ 0.20513561218976975,
863
+ 0.1818600483238697,
864
+ 0.12630234509706498,
865
+ 0.13205429986119271,
866
+ 0.1874783955514431,
867
+ 0.1618017990142107,
868
+ 0.2584618657827377,
869
+ 0.1475326895713806,
870
+ 0.11145277582108974,
871
+ 0.08494718745350838,
872
+ 0.08523617647588252,
873
+ 0.07205090895295144,
874
+ 0.10780602172017098,
875
+ 0.0874978668987751
876
+ ],
877
+ "mean_return": [
878
+ 7.739,
879
+ 15.176000000000005,
880
+ 12.098000000000004,
881
+ 9.485,
882
+ 5.753000000000002,
883
+ 6.2270000000000145,
884
+ 3.3650000000000135,
885
+ -3.6669999999999936,
886
+ 5.1389999999999985,
887
+ 3.878000000000001,
888
+ 4.062000000000003,
889
+ 4.0059999999999985,
890
+ 2.6900000000000053,
891
+ -1.689999999999992,
892
+ -0.17599999999999225,
893
+ -4.028999999999995,
894
+ 3.491000000000014,
895
+ -4.009999999999991,
896
+ -1.9749999999999854,
897
+ -2.0819999999999808,
898
+ -3.917999999999992,
899
+ 1.0160000000000058,
900
+ -0.7449999999999989,
901
+ -3.5979999999999968,
902
+ -3.6459999999999915,
903
+ 0.977000000000008,
904
+ -1.1679999999999933,
905
+ -2.884999999999991,
906
+ -3.520999999999995,
907
+ -1.124999999999995,
908
+ -1.0189999999999926,
909
+ -3.1569999999999947,
910
+ -3.6679999999999877,
911
+ -3.6679999999999957,
912
+ -2.405999999999998,
913
+ -1.1729999999999934,
914
+ 3.932000000000002,
915
+ 3.8420000000000014,
916
+ 1.9799999999999975,
917
+ -1.0560000000000005,
918
+ -0.9879999999999949,
919
+ -3.5929999999999978,
920
+ -1.705999999999986,
921
+ -3.719999999999999,
922
+ 1.8149999999999973,
923
+ -4.013999999999992,
924
+ -1.0299999999999927,
925
+ 1.2470000000000023,
926
+ -2.855999999999998,
927
+ -1.3749999999999925
928
+ ],
929
+ "win_rate_episode": [
930
+ 0.4,
931
+ 0.7,
932
+ 0.6,
933
+ 0.5,
934
+ 0.3,
935
+ 0.4,
936
+ 0.3,
937
+ 0.0,
938
+ 0.3,
939
+ 0.3,
940
+ 0.3,
941
+ 0.3,
942
+ 0.2,
943
+ 0.1,
944
+ 0.1,
945
+ 0.0,
946
+ 0.3,
947
+ 0.0,
948
+ 0.1,
949
+ 0.1,
950
+ 0.0,
951
+ 0.2,
952
+ 0.1,
953
+ 0.0,
954
+ 0.0,
955
+ 0.2,
956
+ 0.1,
957
+ 0.0,
958
+ 0.0,
959
+ 0.1,
960
+ 0.1,
961
+ 0.0,
962
+ 0.0,
963
+ 0.0,
964
+ 0.0,
965
+ 0.1,
966
+ 0.3,
967
+ 0.3,
968
+ 0.2,
969
+ 0.1,
970
+ 0.1,
971
+ 0.0,
972
+ 0.1,
973
+ 0.0,
974
+ 0.2,
975
+ 0.0,
976
+ 0.1,
977
+ 0.2,
978
+ 0.0,
979
+ 0.1
980
+ ],
981
+ "id_winrate": [
982
+ 0.325,
983
+ 0.175,
984
+ 0.125,
985
+ 0.07500000000000001,
986
+ 0.05
987
+ ],
988
+ "id_winrate_iter": [
989
+ 100,
990
+ 200,
991
+ 300,
992
+ 400,
993
+ 500
994
+ ],
995
+ "ood_winrate": [
996
+ 0.0
997
+ ],
998
+ "ood_winrate_iter": [
999
+ 500
1000
+ ],
1001
+ "grad_align_iter": [],
1002
+ "grad_align": [],
1003
+ "rl_grad_norm": [],
1004
+ "bc_grad_norm": [],
1005
+ "repr_drift_iter": [
1006
+ 25,
1007
+ 50,
1008
+ 75,
1009
+ 100,
1010
+ 125,
1011
+ 150,
1012
+ 175,
1013
+ 200,
1014
+ 225,
1015
+ 250,
1016
+ 275,
1017
+ 300,
1018
+ 325,
1019
+ 350,
1020
+ 375,
1021
+ 400,
1022
+ 425,
1023
+ 450,
1024
+ 475,
1025
+ 500
1026
+ ],
1027
+ "repr_drift": [
1028
+ 0.08099153637886047,
1029
+ 0.15414123237133026,
1030
+ 0.26784101128578186,
1031
+ 0.37351346015930176,
1032
+ 0.7241547703742981,
1033
+ 1.372012734413147,
1034
+ 2.410024404525757,
1035
+ 2.198410749435425,
1036
+ 3.3474361896514893,
1037
+ 2.732609987258911,
1038
+ 3.842064380645752,
1039
+ 3.5894956588745117,
1040
+ 4.261956691741943,
1041
+ 3.9535350799560547,
1042
+ 4.4761247634887695,
1043
+ 4.658817291259766,
1044
+ 4.230963706970215,
1045
+ 5.274211883544922,
1046
+ 5.682961463928223,
1047
+ 6.925012111663818
1048
+ ],
1049
+ "t_analysis_iter": [
1050
+ 25,
1051
+ 50,
1052
+ 75,
1053
+ 100,
1054
+ 125,
1055
+ 150,
1056
+ 175,
1057
+ 200,
1058
+ 225,
1059
+ 250,
1060
+ 275,
1061
+ 300,
1062
+ 325,
1063
+ 350,
1064
+ 375,
1065
+ 400,
1066
+ 425,
1067
+ 450,
1068
+ 475,
1069
+ 500
1070
+ ],
1071
+ "norm_low_t": [
1072
+ 0.541771650314331,
1073
+ 2.2737488746643066,
1074
+ 0.7237007021903992,
1075
+ 0.5551862120628357,
1076
+ 0.6337419152259827,
1077
+ 1.6189945936203003,
1078
+ 0.6787412166595459,
1079
+ 1.4094318151474,
1080
+ 0.3824225664138794,
1081
+ 0.2709895372390747,
1082
+ 5.846121311187744,
1083
+ 2.739320993423462,
1084
+ 0.4784424602985382,
1085
+ 0.3508293032646179,
1086
+ 1.2475589513778687,
1087
+ 0.29420673847198486,
1088
+ 0.19785211980342865,
1089
+ 0.3236626386642456,
1090
+ 0.4648295044898987,
1091
+ 0.1427985578775406
1092
+ ],
1093
+ "norm_high_t": [
1094
+ 0.7055990695953369,
1095
+ 2.0272364616394043,
1096
+ 13.79667854309082,
1097
+ 1.0152698755264282,
1098
+ 4.026651859283447,
1099
+ 1.9825714826583862,
1100
+ 0.7544527053833008,
1101
+ 3.5265908241271973,
1102
+ 0.6596992015838623,
1103
+ 0.5065088868141174,
1104
+ 6.615940093994141,
1105
+ 1.2570514678955078,
1106
+ 0.37380844354629517,
1107
+ 1.3679155111312866,
1108
+ 0.6957656741142273,
1109
+ 0.3654467463493347,
1110
+ 0.3122651278972626,
1111
+ 0.4366055428981781,
1112
+ 1.448127269744873,
1113
+ 0.35273078083992004
1114
+ ],
1115
+ "lowhigh_cos": [
1116
+ 0.15398365259170532,
1117
+ 0.22068777680397034,
1118
+ 0.1607862114906311,
1119
+ 0.6298431158065796,
1120
+ -0.07700009644031525,
1121
+ -0.09699821472167969,
1122
+ 0.7008634805679321,
1123
+ 0.3531211018562317,
1124
+ 0.5518617630004883,
1125
+ -0.014294017106294632,
1126
+ -0.2129386067390442,
1127
+ -0.516147792339325,
1128
+ 0.34591352939605713,
1129
+ 0.02539592608809471,
1130
+ 0.08920550346374512,
1131
+ 0.34271037578582764,
1132
+ -0.05921860411763191,
1133
+ -0.2443743199110031,
1134
+ 0.11032988131046295,
1135
+ 0.5413148999214172
1136
+ ]
1137
+ },
1138
+ "bc_on_wins": {
1139
+ "iter": [
1140
+ 10,
1141
+ 20,
1142
+ 30,
1143
+ 40,
1144
+ 50,
1145
+ 60,
1146
+ 70,
1147
+ 80,
1148
+ 90,
1149
+ 100,
1150
+ 110,
1151
+ 120,
1152
+ 130,
1153
+ 140,
1154
+ 150,
1155
+ 160,
1156
+ 170,
1157
+ 180,
1158
+ 190,
1159
+ 200,
1160
+ 210,
1161
+ 220,
1162
+ 230,
1163
+ 240,
1164
+ 250,
1165
+ 260,
1166
+ 270,
1167
+ 280,
1168
+ 290,
1169
+ 300,
1170
+ 310,
1171
+ 320,
1172
+ 330,
1173
+ 340,
1174
+ 350,
1175
+ 360,
1176
+ 370,
1177
+ 380,
1178
+ 390,
1179
+ 400,
1180
+ 410,
1181
+ 420,
1182
+ 430,
1183
+ 440,
1184
+ 450,
1185
+ 460,
1186
+ 470,
1187
+ 480,
1188
+ 490,
1189
+ 500
1190
+ ],
1191
+ "loss": [
1192
+ 0.0,
1193
+ 0.6147490069270134,
1194
+ 0.8275483787059784,
1195
+ 0.746899276971817,
1196
+ 0.47306891083717345,
1197
+ 0.3345955446362495,
1198
+ 0.4720501869916916,
1199
+ 0.4767592802643776,
1200
+ 0.33308539986610414,
1201
+ 0.2699266143143177,
1202
+ 0.29232263639569284,
1203
+ 0.2295227698981762,
1204
+ 0.1925523594021797,
1205
+ 0.19166812300682068,
1206
+ 0.1963917426764965,
1207
+ 0.19072676077485085,
1208
+ 0.20546746701002122,
1209
+ 0.18404317796230316,
1210
+ 0.1437509298324585,
1211
+ 0.299495330452919,
1212
+ 0.34998399317264556,
1213
+ 0.2714237794280052,
1214
+ 0.21747168600559236,
1215
+ 0.22078402191400529,
1216
+ 0.24086840748786925,
1217
+ 0.2355603113770485,
1218
+ 0.2176094949245453,
1219
+ 0.20913921892642975,
1220
+ 0.1819615826010704,
1221
+ 0.15702315866947175,
1222
+ 0.2001718044281006,
1223
+ 0.22251918762922288,
1224
+ 0.19744784608483315,
1225
+ 0.1922704681754112,
1226
+ 0.21251692175865172,
1227
+ 0.1540554977953434,
1228
+ 0.18297502249479294,
1229
+ 0.1609095260500908,
1230
+ 0.1484507754445076,
1231
+ 0.21213592402637005,
1232
+ 0.15481647998094558,
1233
+ 0.1278105281293392,
1234
+ 0.16644979417324066,
1235
+ 0.10792266763746738,
1236
+ 0.1539995450526476,
1237
+ 0.15449323505163193,
1238
+ 0.12793521136045455,
1239
+ 0.16368902325630189,
1240
+ 0.13108288422226905,
1241
+ 0.08190163783729076
1242
+ ],
1243
+ "mean_return": [
1244
+ 9.536000000000008,
1245
+ 7.321000000000003,
1246
+ 11.018000000000004,
1247
+ 12.311000000000002,
1248
+ 6.998000000000005,
1249
+ 1.7940000000000118,
1250
+ 0.9109999999999989,
1251
+ 4.902000000000004,
1252
+ 15.517999999999997,
1253
+ 3.9030000000000085,
1254
+ 3.6680000000000126,
1255
+ 7.838000000000004,
1256
+ 3.2500000000000044,
1257
+ 5.024999999999999,
1258
+ -3.6199999999999948,
1259
+ -3.773999999999989,
1260
+ 6.5530000000000115,
1261
+ -3.1910000000000007,
1262
+ 4.645000000000012,
1263
+ 1.4110000000000027,
1264
+ 5.860000000000004,
1265
+ -2.7789999999999906,
1266
+ 0.795000000000015,
1267
+ -0.11699999999999519,
1268
+ 2.393000000000001,
1269
+ -2.870999999999996,
1270
+ -0.9779999999999953,
1271
+ 3.761000000000005,
1272
+ 7.352000000000006,
1273
+ 3.263000000000017,
1274
+ -0.18899999999998957,
1275
+ 7.492999999999998,
1276
+ 6.823000000000005,
1277
+ 2.5690000000000026,
1278
+ 5.278000000000002,
1279
+ 2.0330000000000057,
1280
+ 5.92800000000001,
1281
+ -1.3679999999999892,
1282
+ 4.517000000000008,
1283
+ -1.1349999999999885,
1284
+ 3.784000000000006,
1285
+ -0.6479999999999938,
1286
+ 0.10600000000000223,
1287
+ 6.8050000000000095,
1288
+ -2.7379999999999947,
1289
+ -0.2789999999999951,
1290
+ 7.730000000000008,
1291
+ 3.2180000000000035,
1292
+ 5.389000000000003,
1293
+ 7.314000000000005
1294
+ ],
1295
+ "win_rate_episode": [
1296
+ 0.5,
1297
+ 0.4,
1298
+ 0.5,
1299
+ 0.6,
1300
+ 0.4,
1301
+ 0.2,
1302
+ 0.1,
1303
+ 0.3,
1304
+ 0.7,
1305
+ 0.3,
1306
+ 0.3,
1307
+ 0.4,
1308
+ 0.2,
1309
+ 0.3,
1310
+ 0.1,
1311
+ 0.0,
1312
+ 0.4,
1313
+ 0.0,
1314
+ 0.3,
1315
+ 0.2,
1316
+ 0.3,
1317
+ 0.0,
1318
+ 0.2,
1319
+ 0.1,
1320
+ 0.2,
1321
+ 0.0,
1322
+ 0.0,
1323
+ 0.3,
1324
+ 0.4,
1325
+ 0.3,
1326
+ 0.1,
1327
+ 0.4,
1328
+ 0.4,
1329
+ 0.2,
1330
+ 0.3,
1331
+ 0.2,
1332
+ 0.4,
1333
+ 0.1,
1334
+ 0.3,
1335
+ 0.1,
1336
+ 0.2,
1337
+ 0.1,
1338
+ 0.1,
1339
+ 0.4,
1340
+ 0.0,
1341
+ 0.1,
1342
+ 0.4,
1343
+ 0.2,
1344
+ 0.3,
1345
+ 0.4
1346
+ ],
1347
+ "id_winrate": [
1348
+ 0.37499999999999994,
1349
+ 0.125,
1350
+ 0.125,
1351
+ 0.07500000000000001,
1352
+ 0.07500000000000001
1353
+ ],
1354
+ "id_winrate_iter": [
1355
+ 100,
1356
+ 200,
1357
+ 300,
1358
+ 400,
1359
+ 500
1360
+ ],
1361
+ "ood_winrate": [
1362
+ 0.0
1363
+ ],
1364
+ "ood_winrate_iter": [
1365
+ 500
1366
+ ],
1367
+ "grad_align_iter": [],
1368
+ "grad_align": [],
1369
+ "rl_grad_norm": [],
1370
+ "bc_grad_norm": [],
1371
+ "repr_drift_iter": [
1372
+ 25,
1373
+ 50,
1374
+ 75,
1375
+ 100,
1376
+ 125,
1377
+ 150,
1378
+ 175,
1379
+ 200,
1380
+ 225,
1381
+ 250,
1382
+ 275,
1383
+ 300,
1384
+ 325,
1385
+ 350,
1386
+ 375,
1387
+ 400,
1388
+ 425,
1389
+ 450,
1390
+ 475,
1391
+ 500
1392
+ ],
1393
+ "repr_drift": [
1394
+ 0.6298390626907349,
1395
+ 1.384177327156067,
1396
+ 2.326510429382324,
1397
+ 2.968867778778076,
1398
+ 4.350086688995361,
1399
+ 5.24165678024292,
1400
+ 4.392853736877441,
1401
+ 4.643966197967529,
1402
+ 3.483157157897949,
1403
+ 4.787506580352783,
1404
+ 4.09738302230835,
1405
+ 5.379372596740723,
1406
+ 5.29456090927124,
1407
+ 4.70228385925293,
1408
+ 5.3411946296691895,
1409
+ 4.312064170837402,
1410
+ 5.200801849365234,
1411
+ 5.95058012008667,
1412
+ 5.202488899230957,
1413
+ 7.007084369659424
1414
+ ],
1415
+ "t_analysis_iter": [
1416
+ 25,
1417
+ 50,
1418
+ 75,
1419
+ 100,
1420
+ 125,
1421
+ 150,
1422
+ 175,
1423
+ 200,
1424
+ 225,
1425
+ 250,
1426
+ 275,
1427
+ 300,
1428
+ 325,
1429
+ 350,
1430
+ 375,
1431
+ 400,
1432
+ 425,
1433
+ 450,
1434
+ 475,
1435
+ 500
1436
+ ],
1437
+ "norm_low_t": [
1438
+ 0.9623110890388489,
1439
+ 1.6318888664245605,
1440
+ 0.979232132434845,
1441
+ 0.6954500675201416,
1442
+ 7.221502304077148,
1443
+ 3.3638436794281006,
1444
+ 2.699252128601074,
1445
+ 11.074406623840332,
1446
+ 0.7769108414649963,
1447
+ 4.699594974517822,
1448
+ 0.5511308312416077,
1449
+ 0.6783273816108704,
1450
+ 2.7332770824432373,
1451
+ 0.9369832277297974,
1452
+ 5.802252292633057,
1453
+ 1.015380859375,
1454
+ 0.4822460412979126,
1455
+ 0.2590118944644928,
1456
+ 2.103935718536377,
1457
+ 13.142251968383789
1458
+ ],
1459
+ "norm_high_t": [
1460
+ 16.76836395263672,
1461
+ 8.49650764465332,
1462
+ 4.285837650299072,
1463
+ 6.7723870277404785,
1464
+ 4.769031047821045,
1465
+ 5.21500825881958,
1466
+ 3.6925623416900635,
1467
+ 7.541281223297119,
1468
+ 3.93727970123291,
1469
+ 3.2928314208984375,
1470
+ 1.8251924514770508,
1471
+ 2.1251378059387207,
1472
+ 0.7079325914382935,
1473
+ 3.325742721557617,
1474
+ 2.185340642929077,
1475
+ 3.9786763191223145,
1476
+ 1.6591521501541138,
1477
+ 1.2101223468780518,
1478
+ 1.0581400394439697,
1479
+ 3.666442394256592
1480
+ ],
1481
+ "lowhigh_cos": [
1482
+ 0.45767301321029663,
1483
+ 0.2987271547317505,
1484
+ -0.0011283881030976772,
1485
+ -0.015486480668187141,
1486
+ -0.02862636372447014,
1487
+ 0.32045242190361023,
1488
+ -0.14974863827228546,
1489
+ 0.21088194847106934,
1490
+ 0.27803245186805725,
1491
+ 0.4605426788330078,
1492
+ 0.5095648169517517,
1493
+ -0.3473619818687439,
1494
+ 0.32257556915283203,
1495
+ 0.21602246165275574,
1496
+ 0.5774956941604614,
1497
+ 0.028346825391054153,
1498
+ 0.04856609180569649,
1499
+ -0.30783888697624207,
1500
+ 0.46236172318458557,
1501
+ 0.12084074318408966
1502
+ ]
1503
+ },
1504
+ "low_t_only": {
1505
+ "iter": [
1506
+ 10,
1507
+ 20,
1508
+ 30,
1509
+ 40,
1510
+ 50,
1511
+ 60,
1512
+ 70,
1513
+ 80,
1514
+ 90,
1515
+ 100,
1516
+ 110,
1517
+ 120,
1518
+ 130,
1519
+ 140,
1520
+ 150,
1521
+ 160,
1522
+ 170,
1523
+ 180,
1524
+ 190,
1525
+ 200,
1526
+ 210,
1527
+ 220,
1528
+ 230,
1529
+ 240,
1530
+ 250,
1531
+ 260,
1532
+ 270,
1533
+ 280,
1534
+ 290,
1535
+ 300,
1536
+ 310,
1537
+ 320,
1538
+ 330,
1539
+ 340,
1540
+ 350,
1541
+ 360,
1542
+ 370,
1543
+ 380,
1544
+ 390,
1545
+ 400,
1546
+ 410,
1547
+ 420,
1548
+ 430,
1549
+ 440,
1550
+ 450,
1551
+ 460,
1552
+ 470,
1553
+ 480,
1554
+ 490,
1555
+ 500
1556
+ ],
1557
+ "loss": [
1558
+ 0.14241273887455463,
1559
+ 0.7457631937228143,
1560
+ 0.3223351809661835,
1561
+ 0.06391186248511076,
1562
+ 0.038416425697505474,
1563
+ 0.17279254915192724,
1564
+ 0.06769378646276891,
1565
+ 0.2095325694506755,
1566
+ 0.009423912805505096,
1567
+ 0.16590883563039824,
1568
+ 0.012637624621856958,
1569
+ 0.0877548771444708,
1570
+ 0.00813584040151909,
1571
+ 0.07707010595477186,
1572
+ 0.012053013269905932,
1573
+ 0.08658939634333365,
1574
+ 0.08423460513586178,
1575
+ 0.04880067176418379,
1576
+ 0.033953274900704856,
1577
+ 0.012992696260334924,
1578
+ 0.008944884140510112,
1579
+ 0.016019769360718784,
1580
+ 0.009055553073994815,
1581
+ 0.01699253911792766,
1582
+ 0.010762585152770043,
1583
+ 0.01595191620872356,
1584
+ 0.00753016233938979,
1585
+ 0.04689169134071562,
1586
+ 0.010906433554191608,
1587
+ 0.056269463570788505,
1588
+ 0.019366807251935826,
1589
+ 0.0062401049741311,
1590
+ 0.12801412070384685,
1591
+ 0.024806116382751497,
1592
+ 0.0007252439007061184,
1593
+ 0.004692662098500477,
1594
+ 0.055812038857220615,
1595
+ 0.008202779742987332,
1596
+ 0.11748617056542798,
1597
+ 0.060509867762857536,
1598
+ 0.005952368767793814,
1599
+ 0.003168158449108205,
1600
+ 0.11608589614589845,
1601
+ 0.02811428838458596,
1602
+ 0.05585574175749457,
1603
+ 0.2106617392924818,
1604
+ 0.0026359208328358362,
1605
+ 0.021516407148419604,
1606
+ 0.01737550176321747,
1607
+ 0.004771639059026711
1608
+ ],
1609
+ "mean_return": [
1610
+ 14.21700000000001,
1611
+ 3.6380000000000075,
1612
+ 9.428000000000003,
1613
+ 3.3620000000000103,
1614
+ 6.908000000000006,
1615
+ 1.692000000000002,
1616
+ 2.1159999999999988,
1617
+ -1.5569999999999915,
1618
+ -4.159999999999986,
1619
+ 1.2480000000000064,
1620
+ -0.4949999999999973,
1621
+ -1.41099999999999,
1622
+ 1.4840000000000066,
1623
+ -3.245999999999995,
1624
+ -0.44900000000000073,
1625
+ -3.377999999999998,
1626
+ 1.3000000000000074,
1627
+ -2.1289999999999862,
1628
+ -3.4419999999999917,
1629
+ -1.1989999999999932,
1630
+ -1.7069999999999852,
1631
+ -3.7619999999999933,
1632
+ -3.445999999999999,
1633
+ 1.5520000000000038,
1634
+ -4.1519999999999895,
1635
+ -3.561999999999992,
1636
+ -0.9549999999999947,
1637
+ -0.36000000000000204,
1638
+ -3.521999999999997,
1639
+ -4.9099999999999815,
1640
+ -3.213999999999989,
1641
+ -4.34599999999998,
1642
+ -3.564000000000001,
1643
+ 1.5640000000000012,
1644
+ -3.1959999999999957,
1645
+ -0.0029999999999990477,
1646
+ -4.325999999999986,
1647
+ -3.403999999999999,
1648
+ 1.496000000000005,
1649
+ -3.745999999999995,
1650
+ -3.8859999999999886,
1651
+ -4.101999999999987,
1652
+ -3.7199999999999998,
1653
+ -4.589999999999986,
1654
+ -1.0049999999999926,
1655
+ 1.148000000000008,
1656
+ -1.1509999999999867,
1657
+ -4.233999999999988,
1658
+ -1.5789999999999917,
1659
+ -3.0820000000000025
1660
+ ],
1661
+ "win_rate_episode": [
1662
+ 0.7,
1663
+ 0.3,
1664
+ 0.5,
1665
+ 0.3,
1666
+ 0.4,
1667
+ 0.2,
1668
+ 0.2,
1669
+ 0.1,
1670
+ 0.0,
1671
+ 0.2,
1672
+ 0.1,
1673
+ 0.1,
1674
+ 0.2,
1675
+ 0.0,
1676
+ 0.1,
1677
+ 0.0,
1678
+ 0.2,
1679
+ 0.1,
1680
+ 0.0,
1681
+ 0.1,
1682
+ 0.1,
1683
+ 0.0,
1684
+ 0.0,
1685
+ 0.2,
1686
+ 0.0,
1687
+ 0.0,
1688
+ 0.1,
1689
+ 0.1,
1690
+ 0.0,
1691
+ 0.0,
1692
+ 0.0,
1693
+ 0.0,
1694
+ 0.0,
1695
+ 0.2,
1696
+ 0.0,
1697
+ 0.1,
1698
+ 0.0,
1699
+ 0.0,
1700
+ 0.2,
1701
+ 0.0,
1702
+ 0.0,
1703
+ 0.0,
1704
+ 0.0,
1705
+ 0.0,
1706
+ 0.1,
1707
+ 0.2,
1708
+ 0.1,
1709
+ 0.0,
1710
+ 0.1,
1711
+ 0.0
1712
+ ],
1713
+ "id_winrate": [
1714
+ 0.1,
1715
+ 0.125,
1716
+ 0.05,
1717
+ 0.07500000000000001,
1718
+ 0.05
1719
+ ],
1720
+ "id_winrate_iter": [
1721
+ 100,
1722
+ 200,
1723
+ 300,
1724
+ 400,
1725
+ 500
1726
+ ],
1727
+ "ood_winrate": [
1728
+ 0.03333333333333333
1729
+ ],
1730
+ "ood_winrate_iter": [
1731
+ 500
1732
+ ],
1733
+ "grad_align_iter": [],
1734
+ "grad_align": [],
1735
+ "rl_grad_norm": [],
1736
+ "bc_grad_norm": [],
1737
+ "repr_drift_iter": [
1738
+ 25,
1739
+ 50,
1740
+ 75,
1741
+ 100,
1742
+ 125,
1743
+ 150,
1744
+ 175,
1745
+ 200,
1746
+ 225,
1747
+ 250,
1748
+ 275,
1749
+ 300,
1750
+ 325,
1751
+ 350,
1752
+ 375,
1753
+ 400,
1754
+ 425,
1755
+ 450,
1756
+ 475,
1757
+ 500
1758
+ ],
1759
+ "repr_drift": [
1760
+ 2.9462854862213135,
1761
+ 4.0175580978393555,
1762
+ 2.643015146255493,
1763
+ 2.551987648010254,
1764
+ 3.312786817550659,
1765
+ 2.4359958171844482,
1766
+ 2.559072732925415,
1767
+ 2.8198697566986084,
1768
+ 2.3854217529296875,
1769
+ 3.6191952228546143,
1770
+ 3.2479796409606934,
1771
+ 1.8640403747558594,
1772
+ 3.854935646057129,
1773
+ 5.600114345550537,
1774
+ 4.419790267944336,
1775
+ 4.976497650146484,
1776
+ 6.506608963012695,
1777
+ 5.592122554779053,
1778
+ 4.9610466957092285,
1779
+ 5.032480716705322
1780
+ ],
1781
+ "t_analysis_iter": [
1782
+ 25,
1783
+ 50,
1784
+ 75,
1785
+ 100,
1786
+ 125,
1787
+ 150,
1788
+ 175,
1789
+ 200,
1790
+ 225,
1791
+ 250,
1792
+ 275,
1793
+ 300,
1794
+ 325,
1795
+ 350,
1796
+ 375,
1797
+ 400,
1798
+ 425,
1799
+ 450,
1800
+ 475,
1801
+ 500
1802
+ ],
1803
+ "norm_low_t": [
1804
+ 0.6796470284461975,
1805
+ 0.44823601841926575,
1806
+ 0.20644712448120117,
1807
+ 0.20466354489326477,
1808
+ 0.20883646607398987,
1809
+ 0.04532072693109512,
1810
+ 10.842145919799805,
1811
+ 0.0022803028114140034,
1812
+ 6.848348617553711,
1813
+ 0.17473728954792023,
1814
+ 0.2673734724521637,
1815
+ 0.022294767200946808,
1816
+ 0.007218019105494022,
1817
+ 0.009707015007734299,
1818
+ 0.005265683401376009,
1819
+ 0.000454531837021932,
1820
+ 0.003278611693531275,
1821
+ 1.453465223312378,
1822
+ 0.03150784596800804,
1823
+ 0.011981611140072346
1824
+ ],
1825
+ "norm_high_t": [
1826
+ 8.541013717651367,
1827
+ 1.5492339134216309,
1828
+ 5.437347412109375,
1829
+ 2.7614333629608154,
1830
+ 0.3078320622444153,
1831
+ 0.4148857295513153,
1832
+ 2.6177570819854736,
1833
+ 0.1941348761320114,
1834
+ 2.0243234634399414,
1835
+ 4.235574722290039,
1836
+ 3.396854877471924,
1837
+ 0.6270498037338257,
1838
+ 1.0882025957107544,
1839
+ 1.689743161201477,
1840
+ 2.5909512042999268,
1841
+ 6.877161979675293,
1842
+ 13.16032886505127,
1843
+ 6.755214691162109,
1844
+ 0.44843706488609314,
1845
+ 4.255799770355225
1846
+ ],
1847
+ "lowhigh_cos": [
1848
+ 0.3415203094482422,
1849
+ -0.2844015657901764,
1850
+ -0.1685301512479782,
1851
+ 0.32842129468917847,
1852
+ 0.2436249852180481,
1853
+ 0.06202510744333267,
1854
+ 0.46803396940231323,
1855
+ 0.45338869094848633,
1856
+ 0.6288841962814331,
1857
+ -0.09758282452821732,
1858
+ -0.37640005350112915,
1859
+ 0.6257611513137817,
1860
+ 0.3929409682750702,
1861
+ 0.6832932233810425,
1862
+ 0.5705606937408447,
1863
+ -0.521347165107727,
1864
+ -0.25697511434555054,
1865
+ -0.15323932468891144,
1866
+ 0.5434814095497131,
1867
+ 0.5631635189056396
1868
+ ]
1869
+ },
1870
+ "synthetic": {
1871
+ "iter": [
1872
+ 10,
1873
+ 20,
1874
+ 30,
1875
+ 40,
1876
+ 50,
1877
+ 60,
1878
+ 70,
1879
+ 80,
1880
+ 90,
1881
+ 100,
1882
+ 110,
1883
+ 120,
1884
+ 130,
1885
+ 140,
1886
+ 150,
1887
+ 160,
1888
+ 170,
1889
+ 180,
1890
+ 190,
1891
+ 200,
1892
+ 210,
1893
+ 220,
1894
+ 230,
1895
+ 240,
1896
+ 250,
1897
+ 260,
1898
+ 270,
1899
+ 280,
1900
+ 290,
1901
+ 300,
1902
+ 310,
1903
+ 320,
1904
+ 330,
1905
+ 340,
1906
+ 350,
1907
+ 360,
1908
+ 370,
1909
+ 380,
1910
+ 390,
1911
+ 400,
1912
+ 410,
1913
+ 420,
1914
+ 430,
1915
+ 440,
1916
+ 450,
1917
+ 460,
1918
+ 470,
1919
+ 480,
1920
+ 490,
1921
+ 500
1922
+ ],
1923
+ "loss": [
1924
+ 0.3504294157028198,
1925
+ 0.1905498057603836,
1926
+ 0.1861506998538971,
1927
+ 0.03813919797539711,
1928
+ 0.036733027547597885,
1929
+ 0.03175725042819977,
1930
+ 0.0202049370855093,
1931
+ 0.028819330036640167,
1932
+ 0.06171034649014473,
1933
+ 0.03453705459833145,
1934
+ 0.03766682371497154,
1935
+ 0.01810506545007229,
1936
+ 0.008473776280879974,
1937
+ 0.005288878921419382,
1938
+ 0.007868476212024689,
1939
+ 0.0027722311206161976,
1940
+ 0.001476715668104589,
1941
+ 0.003148599062114954,
1942
+ 0.0027915467508137226,
1943
+ 0.001739312196150422,
1944
+ 0.004062983673065901,
1945
+ 0.0016188174486160278,
1946
+ 0.0013000894105061889,
1947
+ 0.010563315823674202,
1948
+ 0.006723823491483927,
1949
+ 0.009215778671205044,
1950
+ 0.0025935559533536434,
1951
+ 0.00021198474860284477,
1952
+ 0.005951723083853722,
1953
+ 0.00350300339050591,
1954
+ 0.1793643832206726,
1955
+ 0.01968010887503624,
1956
+ 0.00013353567919693887,
1957
+ 0.0026391444262117147,
1958
+ 0.0010560411028563976,
1959
+ 0.00023656449047848582,
1960
+ 0.06821727752685547,
1961
+ 0.0002130581415258348,
1962
+ 0.044811394065618515,
1963
+ 0.0008450949680991471,
1964
+ 0.030531035736203194,
1965
+ 0.0003115326398983598,
1966
+ 0.0006421205471269786,
1967
+ 0.0003727397706825286,
1968
+ 0.0005728481919504702,
1969
+ 0.02530752494931221,
1970
+ 3.54906587745063e-05,
1971
+ 0.0005603608442470431,
1972
+ 9.057774150278419e-05,
1973
+ 0.01868569850921631
1974
+ ],
1975
+ "mean_return": [],
1976
+ "win_rate_episode": [],
1977
+ "id_winrate": [
1978
+ 0.1,
1979
+ 0.15000000000000002,
1980
+ 0.0,
1981
+ 0.125,
1982
+ 0.05
1983
+ ],
1984
+ "id_winrate_iter": [
1985
+ 100,
1986
+ 200,
1987
+ 300,
1988
+ 400,
1989
+ 500
1990
+ ],
1991
+ "ood_winrate": [],
1992
+ "ood_winrate_iter": [],
1993
+ "grad_align_iter": [],
1994
+ "grad_align": [],
1995
+ "rl_grad_norm": [],
1996
+ "bc_grad_norm": [],
1997
+ "repr_drift_iter": [],
1998
+ "repr_drift": [],
1999
+ "t_analysis_iter": [],
2000
+ "norm_low_t": [],
2001
+ "norm_high_t": [],
2002
+ "lowhigh_cos": []
2003
+ }
2004
+ }
2005
+ }