The GVG-AI Competition

Training Set 2-Player Games 4 (Test Set CIG 2016)

Rank	Username	Country	Points	Avg. Glicko Score
1	MaastCTS2	Netherlands	169	696.31
2	adrienctx	Belgium	126	682.39
3	Number27	Germany	124	716.18
4	sampleOLMCTS Sample	United Kingdom	123	672.64
5	NextLevel	Poland	118	617.14
6	ToVo2	Slovenia	108	698.12
7	CatLinux	United Kingdom	71	509.97
8	random Sample	United Kingdom	66	259.61
9	NUDTQi	China	62	562.07
10	webpigeon	United Kingdom	43	152.45

Egg Hunt:

Game	Username	Country	Points	Glicko Score	Winner % (N / D)	Game Score Avg.	Timesteps Avg.
Egg Hunt	MaastCTS2	Netherlands	25	968.49	50% (696 / 0)	44.67	289
Egg Hunt	Number27	Germany	18	967.99	62% (1279 / 5)	25.76	725
Egg Hunt	ToVo2	Slovenia	15	967.8	45% (388 / 0)	40.54	408
Egg Hunt	adrienctx	Belgium	12	915.18	61% (1183 / 0)	53.18	642
Egg Hunt	NextLevel	Poland	10	868.57	34% (1172 / 0)	27.88	746
Egg Hunt	sampleOLMCTS	United Kingdom	8	846.09	39% (1153 / 1)	30.85	838
Egg Hunt	NUDTQi	China	6	787.63	34% (314 / 0)	35.14	546
Egg Hunt	random	United Kingdom	4	586.93	14% (1203 / 0)	20.03	887
Egg Hunt	CatLinux	United Kingdom	2	51.97	47% (1179 / 65)	-19.5	792
Egg Hunt	webpigeon	United Kingdom	1	0	1% (1752 / 1319)	-750.96	185

Fatty:

Game	Username	Country	Points	Glicko Score	Winner % (N / D)	Game Score Avg.	Timesteps Avg.
Fatty	Number27	Germany	25	816.04	66% (1267 / 3)	11.97	483
Fatty	MaastCTS2	Netherlands	18	701.17	64% (696 / 0)	11.96	589
Fatty	CatLinux	United Kingdom	15	683.26	55% (1177 / 0)	10.09	787
Fatty	adrienctx	Belgium	12	650.42	60% (1178 / 0)	12.08	606
Fatty	ToVo2	Slovenia	10	642.19	56% (388 / 0)	9.9	746
Fatty	sampleOLMCTS	United Kingdom	8	613.95	53% (1151 / 0)	9.68	787
Fatty	NUDTQi	China	6	609.53	50% (314 / 0)	9.81	872
Fatty	NextLevel	Poland	4	447.99	38% (1166 / 1)	6.41	852
Fatty	random	United Kingdom	2	206.78	26% (1206 / 0)	7.62	705
Fatty	webpigeon	United Kingdom	1	154.33	34% (1754 / 179)	-94.63	740

I Saw Santa:

Game	Username	Country	Points	Glicko Score	Winner % (N / D)	Game Score Avg.	Timesteps Avg.
I Saw Santa	adrienctx	Belgium	25	885.56	71% (1182 / 0)	0.71	1238
I Saw Santa	MaastCTS2	Netherlands	18	811.38	55% (696 / 1)	-0.89	1260
I Saw Santa	NextLevel	Poland	15	750.56	53% (1171 / 2)	-1.17	1280
I Saw Santa	sampleOLMCTS	United Kingdom	12	706.42	43% (1154 / 0)	0.43	1404
I Saw Santa	NUDTQi	China	10	700.53	45% (314 / 1)	-2.74	1433
I Saw Santa	ToVo2	Slovenia	8	696.62	47% (388 / 0)	0.47	1355
I Saw Santa	Number27	Germany	6	695.47	47% (1280 / 0)	0.47	1382
I Saw Santa	CatLinux	United Kingdom	4	659.42	44% (1177 / 1)	-0.41	1397
I Saw Santa	random	United Kingdom	2	654.58	42% (1201 / 0)	0.42	1435
I Saw Santa	webpigeon	United Kingdom	1	540.25	37% (1749 / 137)	-77.96	1331

Mimic:

Game	Username	Country	Points	Glicko Score	Winner % (N / D)	Game Score Avg.	Timesteps Avg.
Mimic	NextLevel	Poland	25	940.59	46% (328 / 0)	0.46	902
Mimic	sampleOLMCTS	United Kingdom	18	879.2	54% (965 / 1)	-0.49	981
Mimic	ToVo2	Slovenia	15	878.72	51% (330 / 0)	0.51	933
Mimic	CatLinux	United Kingdom	12	832.24	54% (968 / 0)	0.54	980
Mimic	webpigeon	United Kingdom	10	819.64	54% (1522 / 112)	-73.05	930
Mimic	random	United Kingdom	8	808.19	53% (997 / 0)	0.53	988
Mimic	adrienctx	Belgium	6	807.57	37% (969 / 0)	0.37	982
Mimic	Number27	Germany	4	804.21	53% (1052 / 0)	0.53	987
Mimic	MaastCTS2	Netherlands	2	591.06	31% (594 / 0)	0.31	969

Reflection:

Game	Username	Country	Points	Glicko Score	Winner % (N / D)	Game Score Avg.	Timesteps Avg.
Reflection	MaastCTS2	Netherlands	25	762.52	19% (696 / 0)	0.19	1003
Reflection	sampleOLMCTS	United Kingdom	18	741.36	20% (1153 / 0)	0.2	1129
Reflection	NextLevel	Poland	15	738.58	21% (1187 / 0)	0.21	1027
Reflection	CatLinux	United Kingdom	12	736.42	22% (1176 / 24)	-20.19	1082
Reflection	ToVo2	Slovenia	10	727.44	18% (388 / 0)	0.18	983
Reflection	adrienctx	Belgium	8	688.97	14% (1190 / 3)	-2.38	1019
Reflection	Number27	Germany	6	673.79	19% (1294 / 17)	-12.95	1035
Reflection	NUDTQi	China	4	594.29	18% (314 / 0)	0.18	908
Reflection	random	United Kingdom	2	44.88	2% (1203 / 0)	0.02	259
Reflection	webpigeon	United Kingdom	1	0	3% (1748 / 1480)	-846.66	304

Train Ride:

Game	Username	Country	Points	Glicko Score	Winner % (N / D)	Game Score Avg.	Timesteps Avg.
Train Ride	MaastCTS2	Netherlands	25	1377.17	65% (696 / 0)	42.61	296
Train Ride	Number27	Germany	18	1368.25	60% (1274 / 0)	40.43	261
Train Ride	adrienctx	Belgium	15	1358.82	67% (1173 / 0)	90.69	366
Train Ride	ToVo2	Slovenia	12	1248.15	52% (386 / 0)	41.35	245
Train Ride	CatLinux	United Kingdom	10	1219.57	54% (1175 / 1)	54.05	235
Train Ride	sampleOLMCTS	United Kingdom	8	1199.33	59% (1143 / 0)	37.88	245
Train Ride	NextLevel	Poland	6	1093.68	50% (1107 / 1)	41.18	269
Train Ride	NUDTQi	China	4	1045.68	36% (310 / 0)	44.77	249
Train Ride	random	United Kingdom	2	23.9	6% (1204 / 0)	16.18	123
Train Ride	webpigeon	United Kingdom	1	0	8% (1750 / 824)	-456.65	196

Treasure Map:

Game	Username	Country	Points	Glicko Score	Winner % (N / D)	Game Score Avg.	Timesteps Avg.
Treasure Map	random	United Kingdom	25	0	0% (1135 / 0)	0	1396
Treasure Map	NUDTQi	China	18	0	0% (156 / 1)	-6.41	1084
Treasure Map	adrienctx	Belgium	15	0	0% (1044 / 19)	-18.2	1360
Treasure Map	NextLevel	Poland	12	0	0% (1065 / 78)	-73.24	1107
Treasure Map	sampleOLMCTS	United Kingdom	10	0	0% (1071 / 166)	-155	1323
Treasure Map	MaastCTS2	Netherlands	8	0	0% (576 / 103)	-178.82	1045
Treasure Map	ToVo2	Slovenia	6	0	0% (378 / 72)	-190.48	1260
Treasure Map	Number27	Germany	4	0	0% (1126 / 216)	-191.83	1050
Treasure Map	CatLinux	United Kingdom	2	0	0% (1087 / 237)	-218.03	1230
Treasure Map	webpigeon	United Kingdom	1	0	0% (1712 / 1541)	-900.12	380

Upgrade-X:

Game	Username	Country	Points	Glicko Score	Winner % (N / D)	Game Score Avg.	Timesteps Avg.
Upgrade-X	sampleOLMCTS	United Kingdom	25	500.51	56% (1081 / 2)	135.71	945
Upgrade-X	ToVo2	Slovenia	18	467.39	49% (388 / 0)	132.01	862
Upgrade-X	NextLevel	Poland	15	424.11	42% (1082 / 0)	127.09	864
Upgrade-X	MaastCTS2	Netherlands	12	359.54	38% (628 / 2)	110.24	764
Upgrade-X	Number27	Germany	10	217.11	24% (1189 / 9)	119.99	900
Upgrade-X	NUDTQi	China	8	178.39	26% (314 / 1)	137.59	834
Upgrade-X	adrienctx	Belgium	6	137.76	21% (1093 / 0)	119.88	941
Upgrade-X	random	United Kingdom	4	14.6	6% (1146 / 0)	108.2	352
Upgrade-X	CatLinux	United Kingdom	2	0	25% (1107 / 304)	-177.33	728
Upgrade-X	webpigeon	United Kingdom	1	0	0% (1696 / 1696)	-1000	1

War Zone:

Game	Username	Country	Points	Glicko Score	Winner % (N / D)	Game Score Avg.	Timesteps Avg.
War Zone	Number27	Germany	25	1618.98	69% (1275 / 5)	89.98	351
War Zone	MaastCTS2	Netherlands	18	1381.52	44% (696 / 1)	52.77	562
War Zone	adrienctx	Belgium	15	1379.57	52% (1175 / 1)	42.93	676
War Zone	ToVo2	Slovenia	12	1352.92	44% (387 / 0)	37.78	695
War Zone	sampleOLMCTS	United Kingdom	10	1239.52	36% (1153 / 0)	29.58	810
War Zone	CatLinux	United Kingdom	8	1239.08	45% (1178 / 2)	29.59	833
War Zone	NextLevel	Poland	6	1230.73	39% (1189 / 0)	30.31	721
War Zone	NUDTQi	China	4	1142.61	24% (314 / 0)	9.95	726
War Zone	random	United Kingdom	2	804.8	8% (1205 / 0)	-6.85	526
War Zone	webpigeon	United Kingdom	1	0	2% (1750 / 1580)	-903.08	144

Wheel Me:

Game	Username	Country	Points	Glicko Score	Winner % (N / D)	Game Score Avg.	Timesteps Avg.
Wheel Me	webpigeon	United Kingdom	25	10.27	11% (1753 / 485)	-276.56	1360
Wheel Me	MaastCTS2	Netherlands	18	10.27	17% (696 / 1)	-1.27	1476
Wheel Me	random	United Kingdom	15	0	12% (1205 / 0)	0.12	1710
Wheel Me	adrienctx	Belgium	12	0	19% (1180 / 0)	0.19	1567
Wheel Me	NextLevel	Poland	10	0	18% (1172 / 0)	0.18	1483
Wheel Me	Number27	Germany	8	0	15% (1272 / 1)	-0.64	1606
Wheel Me	sampleOLMCTS	United Kingdom	6	0	13% (1154 / 0)	0.13	1686
Wheel Me	CatLinux	United Kingdom	4	0	12% (1179 / 3)	-2.43	1701
Wheel Me	ToVo2	Slovenia	2	0	11% (388 / 0)	0.11	1718
Wheel Me	NUDTQi	China	1	0	10% (314 / 0)	0.1	1799

Rank	Username	Country	Description	G-1	G-2	G-3	G-4	G-5	G-6	G-7	G-8	G-9	G-10	Total
1	MaastCTS2	Netherlands	Controller description: MaastCTS2 Two-player variant of the MaastCTS2 agent. A number of enhancements from the single-player variant have been disabled. Description	25	18	18	2	25	25	8	12	18	18	169
2	adrienctx	Belgium	Controller description: adrienctx simple MCTS Description	12	12	25	6	8	15	15	6	15	12	126
3	Number27	Germany	Controller description: Number27 Using a GA for local movement and a value map to direct the player across the level. The value map is created by evaluating each object type which results in a certain influence across the map. Notable events, the time spent in one area or non deterministic movements influence the players behaviour or how it chooses its optimal action. Description	18	25	6	4	6	18	4	10	25	8	124
4	sampleOLMCTS Sample	United Kingdom	Controller description: sampleOLMCTS Sample OLMCTS Description	8	8	12	18	18	8	10	25	10	6	123
5	NextLevel	Poland	Controller description: NextLevel OLMCTS equipped with a state heuristic based on features of objects, evaluated during preliminary learning. Description	10	4	15	25	15	6	12	15	6	10	118
6	ToVo2	Slovenia	Controller description: ToVo2 A Sarsa-UCT algorithm, using TD backups and space-local Q-value normalization. Builds a tree, does not compute transpositions. Most primitive opponent model: random player, assumed as part of the environment � keeping value estimates only for own agent. No expert knowledge, no handcrafted strategies, no heuristics. Description	15	10	8	15	10	12	6	18	12	2	108
7	CatLinux	United Kingdom	Controller description: CatLinux RHGA with circular buffer Description	2	15	4	12	12	10	2	2	8	4	71
8	random Sample	United Kingdom	Controller description: random Sample random Description	4	2	2	8	2	2	25	4	2	15	66
9	NUDTQi	China	Controller description: NUDTQi The method is based on 1+1 RHEA, there are following changes: save the history optimal information to shift on next tick; the search depth is initially 1, after finishing the least iterations then increase gradually; the mutation probability is changeable according to the max search depth Description	6	6	10	1	4	4	18	8	4	1	62
10	webpigeon	United Kingdom	Controller description: webpigeon OLMCTS with simple agent prediction Description	1	1	1	10	1	1	1	1	1	25	43

* Note: These rankings only include controllers executed after 01/05/2016

* For details on how rankings are calculated, check the GVGAI Wiki.