0

0

I'm trying to reproduce the work in the paper Demand Response for Home Energy Management Using Reinforcement Learning and Artificial Neural Network. I want to optimize the power consumption for home appliances. The action space is a different power rating for home appliances. My reward function is = -(power rating *electricity price).

I have trained an RL agent using DQN algorithm on Matlab. I have action space that the agent should select from, but my agent always takes the same action irrespective of state. I have checked my reward function and the algorithm does not select the action with the highest reward. Anyone can think of why is the agent behaving this way?

My code:

enter image description here enter image description here What I'm getting while training:

enter image description here

And my agent always takes the same power rating regardless of the state (electricity price). Why?

Aya
  • 1
  • 2
  • Please add more detail of what you have tried, e.g. a snippet from your code, expected results vs actual results, etc. The problem could caused by any number of reason, and without more information I just can't say. – mimocha Jul 26 '20 at 14:50
  • I have edit the script. – Aya Jul 26 '20 at 16:27
  • If you think this is not a bug, but a conceptual issue, maybe this question can be asked on [Artificial Intelligence Stack Exchange](https://ai.stackexchange.com/), which is the most appropriate site to ask theoretical questions related to reinforcement learning. If you ask it there, please, delete it from here (to avoid cross-posting, which is generally discouraged). If you think this is a bug in your code, then just leave this question here. However, if you didn't know it, now you know that there's AI SE to ask theoretical RL questions :) – nbro Jul 26 '20 at 22:15

0 Answers0