Actor$-$Critic Problem:

$$ Design the Actor$-$Critic algorithm using TensorFlow.

$$ Design Reward Function.

$$ Environment Solution

$$ Train the model over $500$ episodes to minimize energy consumption while

maintaining an indoor temperature of $22\backslash$deg C$.$

$$ Evaluate the performance of the model on test set to measure its performance

$$ Provide graphs showing the convergence of the Actor and Critic losses.

$$ Plot the learned policy by showing the action probabilities across different state

values $($e$.$g$.,$ temperature settings$).$

$$ Provide an analysis on a comparison of the energy consumption before and

after applying the reinforcement learning algorithm.