Is there a machine learning model that can be trained with labels that only say how "right" or "wrong" it was?
What are the major differences between multi-armed bandits and the other well-known algorithms (DQN, A3C, PPO, etc)?
Mathematically, what is happening differently in the neural net during exploration vs. exploitation?
When using experience replay, do we update the parameters for all samples of the mini-batch or for each sample in the mini-batch separately?
How can you represent the state and action spaces for a card game in the case of a variable number of cards and actions?
Only top scored, non community-wiki answers of a minimum length are eligible