Atari Games On Atari 2600 Tutankham

Score

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle	Score	Paper Title	Repository
DQN hs	45.6	Deep Reinforcement Learning with Double Q-learning
SARSA	98.2	-	-
DDQN (tuned) noop	218.4	Dueling Network Architectures for Deep Reinforcement Learning
Prior noop	204.6	Prioritized Experience Replay
GDI-I3	423.9	Generalized Data Distribution Iteration	-
R2D2	395.3	Recurrent Experience Replay in Distributed Reinforcement Learning	-
MuZero	491.48	Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model
Prior+Duel noop	245.9	Dueling Network Architectures for Deep Reinforcement Learning
NoisyNet-Dueling	269	Noisy Networks for Exploration
CGP	0	Evolving simple programs for playing Atari games
DDQN+Pop-Art noop	183.9	Learning values across many orders of magnitude	-
GDI-I3	423.9	GDI: Rethinking What Makes Reinforcement Learning Different From Supervised Learning	-
DDQN (tuned) hs	92.2	Deep Reinforcement Learning with Double Q-learning
A3C LSTM hs	144.2	Asynchronous Methods for Deep Reinforcement Learning
GDI-H3	418.2	Generalized Data Distribution Iteration	-
C51 noop	280.0	A Distributional Perspective on Reinforcement Learning
Advantage Learning	245.22	Increasing the Action Gap: New Operators for Reinforcement Learning
DQN noop	68.1	Deep Reinforcement Learning with Double Q-learning
Best Learner	114.3	The Arcade Learning Environment: An Evaluation Platform for General Agents
ASL DDQN	252.9	Train a Real-world Local Path Planner in One Hour via Partially Decoupled Reinforcement Learning and Vectorized Diversity	-

0 of 44 row(s) selected.