Atari Games On Atari 2600 Breakout

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

		Paper Title	Repository
GDI-I3(200M frames)	864.00	Generalized Data Distribution Iteration	-
GDI-H3(200M frames)	864.00	Generalized Data Distribution Iteration	-
GDI-H3	864	Generalized Data Distribution Iteration	-
GDI-I3	864.00	Generalized Data Distribution Iteration	-
Bootstrapped DQN	855	Deep Exploration via Bootstrapped DQN
FQF	854.2	Fully Parameterized Quantile Function for Distributional Reinforcement Learning
R2D2	837.7	Recurrent Experience Replay in Distributed Reinforcement Learning	-
Ape-X	800.9	Distributed Prioritized Experience Replay
Agent57	790.4	Agent57: Outperforming the Atari Human Benchmark
IMPALA (deep)	787.34	IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures
A3C LSTM hs	766.8	Asynchronous Methods for Deep Reinforcement Learning
MuZero (Res2 Adam)	758.04	Online and Offline Reinforcement Learning by Planning with a Learned Model
C51 noop	748.0	A Distributional Perspective on Reinforcement Learning
QR-DQN-1	742	Distributional Reinforcement Learning with Quantile Regression
IQN	734	Implicit Quantile Networks for Distributional Reinforcement Learning
A3C FF hs	681.9	Asynchronous Methods for Deep Reinforcement Learning
DNA	626	DNA: Proximal Policy Optimization with a Dual Network Architecture
ASL DDQN	621.7	Train a Real-world Local Path Planner in One Hour via Partially Decoupled Reinforcement Learning and Vectorized Diversity
A3C FF (1 day) hs	551.6	Asynchronous Methods for Deep Reinforcement Learning
Reactor 500M	514.8	The Reactor: A fast and sample-efficient Actor-Critic agent for Reinforcement Learning	-

0 of 58 row(s) selected.