Atari Games On Atari 2600 Up And Down

评估指标

Score

评测结果

各个模型在此基准测试上的表现结果

模型名称	Score	Paper Title	Repository
QR-DQN-1	71260	Distributional Reinforcement Learning with Quantile Regression	-
Best Learner	3532.7	The Arcade Learning Environment: An Evaluation Platform for General Agents	-
ES FF (1 hour) noop	67974.0	Evolution Strategies as a Scalable Alternative to Reinforcement Learning	-
SAC	250.7	Soft Actor-Critic for Discrete Action Settings	-
DQN noop	9989.9	Deep Reinforcement Learning with Double Q-learning	-
IQN	88148	Implicit Quantile Networks for Distributional Reinforcement Learning	-
DDQN (tuned) noop	22972.2	Dueling Network Architectures for Deep Reinforcement Learning	-
A3C FF hs	74705.7	Asynchronous Methods for Deep Reinforcement Learning	-
MuZero	715545.61	Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model	-
POP3D	242701.51	Policy Optimization With Penalized Point Probability Distance: An Alternative To Proximal Policy Optimization	-
IMPALA (deep)	332546.75	IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures	-
Nature DQN	8456.0	Human level control through deep reinforcement learning
A3C FF (1 day) hs	54525.4	Asynchronous Methods for Deep Reinforcement Learning	-
A3C LSTM hs	105728.7	Asynchronous Methods for Deep Reinforcement Learning	-
A2C + SIL	53314.6	Self-Imitation Learning	-
DreamerV2	653662	Mastering Atari with Discrete World Models	-
Advantage Learning	13909.74	Increasing the Action Gap: New Operators for Reinforcement Learning	-
CGP	14524	Evolving simple programs for playing Atari games	-
Duel noop	44939.6	Dueling Network Architectures for Deep Reinforcement Learning	-
R2D2	589226.9	Recurrent Experience Replay in Distributed Reinforcement Learning	-

0 of 44 row(s) selected.