HyperAI超神経
ホームニュース最新論文チュートリアルデータセット百科事典SOTALLMモデルGPU ランキング学会
検索
サイトについて
HyperAI超神経

Atari Games On Atari 2600 Kangaroo

評価指標

Score

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

モデル名
Score
Paper TitleRepository
NoisyNet-Dueling15227Noisy Networks for Exploration
QR-DQN-115356Distributional Reinforcement Learning with Quantile Regression
GDI-I314500GDI: Rethinking What Makes Reinforcement Learning Different From Supervised Learning-
Ape-X1416Distributed Prioritized Experience Replay
MuZero16763.60Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model
DQN hs4496.0Deep Reinforcement Learning with Double Q-learning
Nature DQN6740.0Human level control through deep reinforcement learning
MuZero (Res2 Adam)13838Online and Offline Reinforcement Learning by Planning with a Learned Model
Advantage Learning10809.16Increasing the Action Gap: New Operators for Reinforcement Learning
IQN15487Implicit Quantile Networks for Distributional Reinforcement Learning
Prior noop16200.0Prioritized Experience Replay
Prior+Duel noop1792.0Dueling Network Architectures for Deep Reinforcement Learning
Agent5724034.16Agent57: Outperforming the Atari Human Benchmark
ASL DDQN13027Train a Real-world Local Path Planner in One Hour via Partially Decoupled Reinforcement Learning and Vectorized Diversity-
SARSA8.8--
A3C FF hs94.0Asynchronous Methods for Deep Reinforcement Learning
ES FF (1 hour) noop11200.0Evolution Strategies as a Scalable Alternative to Reinforcement Learning
DDQN (tuned) hs11204.0Deep Reinforcement Learning with Double Q-learning
DDQN (tuned) noop12992.0Dueling Network Architectures for Deep Reinforcement Learning
A3C FF (1 day) hs106.0Asynchronous Methods for Deep Reinforcement Learning
0 of 47 row(s) selected.