Dueling Network Architectures for Deep Reinforcement Learning

In recent years there have been many successes of using deep representations in reinforcement learning. Still, many of these applications use conventional architectures, such as convolutional networks, LSTMs, or auto-encoders... (read more)

PDF Abstract

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Atari Games Atari 2600 Alien Duel hs Score 1486.5 # 27
Atari Games Atari 2600 Alien Prior+Duel hs Score 823.7 # 35
Atari Games Atari 2600 Alien DDQN (tuned) noop Score 3747.7 # 18
Atari Games Atari 2600 Alien Duel noop Score 4461.4 # 15
Atari Games Atari 2600 Alien Prior+Duel noop Score 3941.0 # 17