Implementation of DQN and modified DQN on a custom 2D track reports the modified version achieving around 40 average reward after 1000 episodes, 60% higher than standard DQN.
sumo - SUMO Documentation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Enhanced Deep Q-Learning for 2D Self-Driving Cars: Implementation and Evaluation on a Custom Track Environment
Implementation of DQN and modified DQN on a custom 2D track reports the modified version achieving around 40 average reward after 1000 episodes, 60% higher than standard DQN.