Gabizon
  • Member for 2 years, 5 months
  • Last seen more than 1 year ago
A2C Critic Loss Interpretation
1 votes

Unfortunately no, the way to go is track the total reward and see if it's increasing and converging eventually. Value loss isn't a useful metric as the loss can be 0 when the value network always ...

View answer
What could be the cause of the drop in the reward in A3C?
0 votes

I'll share my understanding so far. This kind of behavior is actually normal when using on-policy algorithms with a sparse final reward. Issue stems from the fact that once you get stuck in a behavior ...

View answer
How to properly optimize shared network between actor and critic?
0 votes

Just use one class inheriting from nn.Module called e.g. ActorCriticModel. Then, have two members called self.actor and self.critic and define them to have the desired architecture.Then, in the ...

View answer