QLearner
An implementation of the Q-learning Algorithm
- based on the explanation provided in the book of Ertel: Introduction to Artificial Intelligence
- Instead of randomly selecting a state s and selecting/carrying out actions according to this random choice, one or more path(s) is/are specified.
- No time-limits / convergence checks are integrated into the implementation. The algorithm iterates a pre-specified number of times over the aforementioned paths.