algorithms.qsom

The Q-SOM learning algorithm, leveraging 2 SOMs and a Q-Table.

The Self-Organizing Maps (SOMs) are used to handle the continuous and multi-dimensional state and action spaces, whereas the Q-Table learns the interests of actions in states.

members:

Modules

qsom	This module defines a Q-SOM helper that is used as an entrypoint to simplify the instantiation of Q-SOM Agents from a Gym Environment.
qsom_agent	This module implements a Q-SOM Agent, with the decision and learning algorithms that make the agent act based on the received observations from the environment.
som	This module defines Self-Organizing Maps, also known as Kohonen Maps.