Atari Games On Atari 2600 Montezumas Revenge

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
R2D2	2061.3	Recurrent Experience Replay in Distributed Reinforcement Learning	-
GDI-H3	2500	Generalized Data Distribution Iteration	-
Best Learner	10.7	The Arcade Learning Environment: An Evaluation Platform for General Agents
MP-EB	142	Incentivizing Exploration In Reinforcement Learning With Deep Predictive Models
Sarsa-φ-EB	2745.4	Count-Based Exploration in Feature Space for Reinforcement Learning
DNA	0	DNA: Proximal Policy Optimization with a Dual Network Architecture
DQN+SR	1778.8	Count-Based Exploration with the Successor Representation
A3C FF (1 day) hs	53	Asynchronous Methods for Deep Reinforcement Learning
IMPALA (deep)	0.00	IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures
Prior+Duel hs	24.0	Deep Reinforcement Learning with Double Q-learning
A3C FF hs	67	Asynchronous Methods for Deep Reinforcement Learning
GDI-I3	3000	GDI: Rethinking What Makes Reinforcement Learning Different From Supervised Learning	-
A2C+CoEX	6635	Contingency-Aware Exploration in Reinforcement Learning	-
Advantage Learning	0.42	Increasing the Action Gap: New Operators for Reinforcement Learning
POP3D	0	Policy Optimization With Penalized Point Probability Distance: An Alternative To Proximal Policy Optimization
Nature DQN	0	Human level control through deep reinforcement learning
DDQN-PC	3459	Unifying Count-Based Exploration and Intrinsic Motivation
ASL DDQN	0	Train a Real-world Local Path Planner in One Hour via Partially Decoupled Reinforcement Learning and Vectorized Diversity	-
Gorila	84	Massively Parallel Methods for Deep Reinforcement Learning
Sarsa-ε	399.5	Count-Based Exploration in Feature Space for Reinforcement Learning

0 of 50 row(s) selected.