Tech moves fast! Stay ahead of the curve with Techopedia!
Join nearly 200,000 subscribers who receive actionable tech insights from Techopedia.
Q-learning is a term for an algorithm structure representing model-free reinforcement learning. By evaluating policy and using stochastic modeling, Q-learning finds the best path forward in a Markov decision process.
The technical makeup of the Q-learning algorithm involves an agent, a set of states and a set of actions per state.
The Q function uses weights for various steps in conjunction with a discount factor in order to value rewards.
Although it may seem like a simple idea, Q-learning is of paramount importance in many types of reinforcement learning and deep learning models. One of the best examples is where deep Q-learning is used to help machine learning programs to learn game-play strategies in various types of video games, for example, in Atari games from the 1980s. Here a convolutional neural network takes samples of game-play in order to work up a stochastic model that will help the computer know how to play the game better over time.
Q-learning has abundant potential for helping to advance artificial intelligence and machine learning.