Hugging Face's logo

GOVINDFROM
/
MindGamesColonelBlutto

Reinforcement Learning
Safetensors
game-theory
colonel-blotto
neurips-2025
graph-neural-networks
preference-learning
llm-distillation
meta-learning
Model card Files
xet
Community
GOVINDFROM commited on
Commit
00011f2
·
verified ·
1 Parent(s): 626f164

Upload eval_scripted_after_ppo.json

Browse files
Files changed (1) hide show
  1. eval_scripted_after_ppo.json +188 -0
eval_scripted_after_ppo.json ADDED
@@ -0,0 +1,188 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "overall": {
3
+ "n_games": 2100,
4
+ "n_wins": 1927,
5
+ "n_draws": 105,
6
+ "n_losses": 68,
7
+ "win_rate": 91.76190476190477,
8
+ "draw_rate": 5.0,
9
+ "loss_rate": 3.2380952380952377,
10
+ "win_rate_ci_95": {
11
+ "lower": 90.50831835820999,
12
+ "upper": 92.8629828701251
13
+ },
14
+ "average_outcome": 0.9426190476190476,
15
+ "outcome_std": 0.2039322894610053,
16
+ "avg_score_player": 2.0785714285714287,
17
+ "avg_score_opponent": 0.22666666666666666,
18
+ "avg_score_differential": 1.851904761904762,
19
+ "score_differential_std": 1.2225933367723631,
20
+ "max_score_differential": 3.0,
21
+ "min_score_differential": -3.0,
22
+ "avg_victory_margin": 2.0768033212247015,
23
+ "avg_defeat_margin": -1.661764705882353
24
+ },
25
+ "per_opponent": {
26
+ "Random": {
27
+ "n_games": 300,
28
+ "n_wins": 192,
29
+ "n_draws": 43,
30
+ "n_losses": 65,
31
+ "win_rate": 64.0,
32
+ "draw_rate": 14.333333333333334,
33
+ "loss_rate": 21.666666666666668,
34
+ "win_rate_ci_95": {
35
+ "lower": 58.422930323031295,
36
+ "upper": 69.22306648268142
37
+ },
38
+ "average_outcome": 0.7116666666666667,
39
+ "outcome_std": 0.41153844156881486,
40
+ "avg_score_player": 2.31,
41
+ "avg_score_opponent": 1.2933333333333332,
42
+ "avg_score_differential": 1.0166666666666666,
43
+ "score_differential_std": 1.76154919192044,
44
+ "max_score_differential": 3.0,
45
+ "min_score_differential": -3.0,
46
+ "avg_victory_margin": 2.1614583333333335,
47
+ "avg_defeat_margin": -1.6923076923076923
48
+ },
49
+ "Balanced": {
50
+ "n_games": 300,
51
+ "n_wins": 300,
52
+ "n_draws": 0,
53
+ "n_losses": 0,
54
+ "win_rate": 100.0,
55
+ "draw_rate": 0.0,
56
+ "loss_rate": 0.0,
57
+ "win_rate_ci_95": {
58
+ "lower": 98.73570287754538,
59
+ "upper": 99.99999999999997
60
+ },
61
+ "average_outcome": 1.0,
62
+ "outcome_std": 0.0,
63
+ "avg_score_player": 3.0,
64
+ "avg_score_opponent": 0.0,
65
+ "avg_score_differential": 3.0,
66
+ "score_differential_std": 0.0,
67
+ "max_score_differential": 3.0,
68
+ "min_score_differential": 3.0,
69
+ "avg_victory_margin": 3.0,
70
+ "avg_defeat_margin": 0.0
71
+ },
72
+ "AntiLast": {
73
+ "n_games": 300,
74
+ "n_wins": 300,
75
+ "n_draws": 0,
76
+ "n_losses": 0,
77
+ "win_rate": 100.0,
78
+ "draw_rate": 0.0,
79
+ "loss_rate": 0.0,
80
+ "win_rate_ci_95": {
81
+ "lower": 98.73570287754538,
82
+ "upper": 99.99999999999997
83
+ },
84
+ "average_outcome": 1.0,
85
+ "outcome_std": 0.0,
86
+ "avg_score_player": 3.0,
87
+ "avg_score_opponent": 0.0,
88
+ "avg_score_differential": 3.0,
89
+ "score_differential_std": 0.0,
90
+ "max_score_differential": 3.0,
91
+ "min_score_differential": 3.0,
92
+ "avg_victory_margin": 3.0,
93
+ "avg_defeat_margin": 0.0
94
+ },
95
+ "NoisedMirror": {
96
+ "n_games": 300,
97
+ "n_wins": 235,
98
+ "n_draws": 62,
99
+ "n_losses": 3,
100
+ "win_rate": 78.33333333333333,
101
+ "draw_rate": 20.666666666666668,
102
+ "loss_rate": 1.0,
103
+ "win_rate_ci_95": {
104
+ "lower": 73.32901513759226,
105
+ "upper": 82.62121649301679
106
+ },
107
+ "average_outcome": 0.8866666666666667,
108
+ "outcome_std": 0.22095751225568733,
109
+ "avg_score_player": 1.24,
110
+ "avg_score_opponent": 0.29333333333333333,
111
+ "avg_score_differential": 0.9466666666666667,
112
+ "score_differential_std": 0.6611774009715966,
113
+ "max_score_differential": 3.0,
114
+ "min_score_differential": -1.0,
115
+ "avg_victory_margin": 1.221276595744681,
116
+ "avg_defeat_margin": -1.0
117
+ },
118
+ "Heavy_0": {
119
+ "n_games": 300,
120
+ "n_wins": 300,
121
+ "n_draws": 0,
122
+ "n_losses": 0,
123
+ "win_rate": 100.0,
124
+ "draw_rate": 0.0,
125
+ "loss_rate": 0.0,
126
+ "win_rate_ci_95": {
127
+ "lower": 98.73570287754538,
128
+ "upper": 99.99999999999997
129
+ },
130
+ "average_outcome": 1.0,
131
+ "outcome_std": 0.0,
132
+ "avg_score_player": 1.0,
133
+ "avg_score_opponent": 0.0,
134
+ "avg_score_differential": 1.0,
135
+ "score_differential_std": 0.0,
136
+ "max_score_differential": 1.0,
137
+ "min_score_differential": 1.0,
138
+ "avg_victory_margin": 1.0,
139
+ "avg_defeat_margin": 0.0
140
+ },
141
+ "Heavy_1": {
142
+ "n_games": 300,
143
+ "n_wins": 300,
144
+ "n_draws": 0,
145
+ "n_losses": 0,
146
+ "win_rate": 100.0,
147
+ "draw_rate": 0.0,
148
+ "loss_rate": 0.0,
149
+ "win_rate_ci_95": {
150
+ "lower": 98.73570287754538,
151
+ "upper": 99.99999999999997
152
+ },
153
+ "average_outcome": 1.0,
154
+ "outcome_std": 0.0,
155
+ "avg_score_player": 1.0,
156
+ "avg_score_opponent": 0.0,
157
+ "avg_score_differential": 1.0,
158
+ "score_differential_std": 0.0,
159
+ "max_score_differential": 1.0,
160
+ "min_score_differential": 1.0,
161
+ "avg_victory_margin": 1.0,
162
+ "avg_defeat_margin": 0.0
163
+ },
164
+ "Heavy_2": {
165
+ "n_games": 300,
166
+ "n_wins": 300,
167
+ "n_draws": 0,
168
+ "n_losses": 0,
169
+ "win_rate": 100.0,
170
+ "draw_rate": 0.0,
171
+ "loss_rate": 0.0,
172
+ "win_rate_ci_95": {
173
+ "lower": 98.73570287754538,
174
+ "upper": 99.99999999999997
175
+ },
176
+ "average_outcome": 1.0,
177
+ "outcome_std": 0.0,
178
+ "avg_score_player": 3.0,
179
+ "avg_score_opponent": 0.0,
180
+ "avg_score_differential": 3.0,
181
+ "score_differential_std": 0.0,
182
+ "max_score_differential": 3.0,
183
+ "min_score_differential": 3.0,
184
+ "avg_victory_margin": 3.0,
185
+ "avg_defeat_margin": 0.0
186
+ }
187
+ }
188
+ }