Our new X account is live! Follow @wizwand_team for updates

Reinforcement Learning on Atari 2600 Games Breakdown

1,399,753Avg Reward (baseline)

PPO with RUDDER

Updated 4d ago

Evaluation Results

Method
PPO with RUDDER 2018.06	1,399,753	1,655,464	18.3	3,568,513	3,641,583	2
PPO with RUDDER 2018.06	125,249	106,076	-15.3	140,080	132,480	-5.4
PPO with RUDDER 2018.06	49,377	51,984	5.3	66,883	78,460	17.3
PPO with RUDDER 2018.06	46,707	60,577	29.7	109,083	178,438	63.6
PPO with RUDDER 2018.06	28,684	46,119	60.8	464,151	400,370	-13.7
PPO with RUDDER 2018.06	25,162	23,029	-8.5	53,000	36,383	-31.4
PPO with RUDDER 2018.06	22,746	12,162	-46.5	32,383	15,068	-53.5
PPO with RUDDER 2018.06	22,065	4,905	-77.8	310,505	154,479	-50.2
PPO with RUDDER 2018.06	20,314	27,992	37.8	35,353	36,717	3.9
PPO with RUDDER 2018.06	19,738	19,196	-2.7	15,248	16,836	10.4
PPO with RUDDER 2018.06	15,013	39,247	161	46,758	190,123	307
PPO with RUDDER 2018.06	12,870	17,023	32.3	24,667	28,067	13.8
PPO with RUDDER 2018.06	12,522	8,290	-33.8	28,763	16,631	-42.2
PPO with RUDDER 2018.06	10,554	18,054	71.1	29,513	102,930	249
PPO with RUDDER 2018.06	9,029	8,027	-11.1	10,368	8,202	-20.9
PPO with RUDDER 2018.06	8,390	10,545	25.7	10,962	17,242	57.3
PPO with RUDDER 2018.06	8,126	4,090	-49.7	102,916	23,367	-77.3
PPO with RUDDER 2018.06	6,900	7,498	8.7	12,120	10,613	-12.4
PPO with RUDDER 2018.06	6,346	29,016	357	18,380	62,593	241
PPO with RUDDER 2018.06	6,183	5,244	-15.2	10,817	9,573	-11.5
PPO with RUDDER 2018.06	5,788	4,242	-26.7	10,373	11,305	9
PPO with RUDDER 2018.06	4,096	5,005	22.2	6,446	6,984	8.3
PPO with RUDDER 2018.06	3,861	3,024	-21.7	6,480	5,950	-8.2
PPO with RUDDER 2018.06	3,790	4,208	11	4,533	5,563	22.7
PPO with RUDDER 2018.06	3,172	4,770	50.4	5,092	7,439	46.1
PPO with RUDDER 2018.06	2,755	9,764	254	5,360	13,500	152
PPO with RUDDER 2018.06	2,372	4,506	89.9	3,994	6,742	68.8
PPO with RUDDER 2018.06	1,878	3,087	64.4	3,218	5,703	77.3
PPO with RUDDER 2018.06	1,393	1,918	37.7	616	1,827	197
PPO with RUDDER 2018.06	1,261	1,341	6.4	1,930	2,092	8.4
PPO with RUDDER 2018.06	1,204	1,415	17.5	1,838	2,233	21.5
PPO with RUDDER 2018.06	1,105	2,462	123	1,616	4,770	195
PPO with RUDDER 2018.06	936	1,194	27.5	1,078	1,335	23.8
PPO with RUDDER 2018.06	787	724	-8	1,242	1,054	-15.1
PPO with RUDDER 2018.06	778	1,106	42.1	1,281	1,860	45.2
PPO with RUDDER 2018.06	759	777	2.5	2,201	1,339	-39.2
PPO with RUDDER 2018.06	738	885	20.1	820	1,350	64.6
PPO with RUDDER 2018.06	217	153	-29.5	430	352	-18.1
PPO with RUDDER 2018.06	123	151	22.7	140	163	16.3
PPO with RUDDER 2018.06	102	34.1	-66.4	100	33.3	-66.7
PPO with RUDDER 2018.06	98	94.7	-3.4	100	99.5	-0.5
PPO with RUDDER 2018.06	61.5	179	191	56.3	192	241
PPO with RUDDER 2018.06	26.7	25.4	-4.8	32	31.4	-1.9
PPO with RUDDER 2018.06	24.9	32.7	31.3	32.2	47.3	46.9
PPO with RUDDER 2018.06	19.5	11.7	-39.9	52	36.3	-30.3
PPO with RUDDER 2018.06	19.2	18.5	-3.9	21	21	0
PPO with RUDDER 2018.06	0	0	38.4	0	0	0
PPO with RUDDER 2018.06	-3.1	-1.9	39.4	-1.4	1	171
PPO with RUDDER 2018.06	-8.4	-5.5	34	-75	0	100
PPO with RUDDER 2018.06	-9.2	-13.1	-41.7	-0.3	-5.1	-1,825
PPO with RUDDER 2018.06	-13.5	-13.5	0.2	-4	-5.3	-32.8
PPO with RUDDER 2018.06	-29,501	-29,911	-1.4	-29,977	-29,978	0