In order to realize on the main requirements and modifications to be done in the code, we aim to implement a basic learning algorithm based on the \epsilon-greedy action-selection strategy. This taks includes:
Modules to introduce algorithm's specific information
Generation of actions according to the allowed parameters
Generation of rewards according to different metrics (to be inputed by the user, as well)
Determine the action-selection procedure according to the desired method (in this case, for \epsilon-greedy)
In order to realize on the main requirements and modifications to be done in the code, we aim to implement a basic learning algorithm based on the \epsilon-greedy action-selection strategy. This taks includes: