After many research, I still can't find a neat answer about this question:
- When I found the loss of my state-action pair. I'm only backpropagating that loss true the network and setting all other output nodes we didn't took an action from with loss = 0. is this right?
- I'm trying to understand which activation I should use in my hidden layers. I'm inputing numbers between 0-10 and outputting a raw value where no activation function should be applied (this is because we calculate the loss with raw values).
Thnx for helping out!