I'm training an RL agent using the DQN algorithm to do a specific task. The environment is represented by a list of $10$ integer numbers from $0$ to $20$. An example would be $[5, 15, 8, 8, 0, \dots]$.

Is it okay to pass the list as floats to the dense layer, or would that impede the learning process? What is the right way to go about passing integer numbers to the neural network?

  • $\begingroup$ It looks like 10 numbers, it's not that much as input for dense layers. I would recommend normalizing them either to float in the range from 0 to 1 or [-1;1] depending on data specifics. $\endgroup$ – Stepan Novikov Nov 16 '20 at 12:56
  • $\begingroup$ This seems to help my model's performance! $\endgroup$ – mark mark Nov 17 '20 at 0:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.