jon
Reputation: 11
Understanding the reward functionality in Reinforcment learning (atari breakout)
I'm trying to understand the reward functionality in Breakout atari implemented by Deepmind. I'm a little confused about the reward. They represent every state using four frames and depending on that the reward for every action will be received after four frames. My question is, what if the ball got stuck, where it receives a lot of rewards how to determine that the same action which is rewarded is the cause of this more reward?
The case I'm talking about:
Upvotes: 1
Views: 311
Answers (0)
Related Questions
- Unbounded increase in Q-Value, consequence of recurrent reward after repeating the same action in Q-Learning
- iterations and reward in q-learning
- Q-Learning in combination with neural-networks (rewarding understanding)
- confusion in selecting reward in q-learning
- What's the principle to design the reward function, of DQN?
- Atari score vs reward in rllib DQN implementation
- Formulation of a reward structure
- How does neural network know which reward it got from action?
- Reward value calculation: Q-Learning
- Qlearning and indexing of reward