Reinforcement Learning: SARSA and Q-Learning — Part 3

Author(s): Tan Pengshi Alvin

Originally published on Towards AI.

Introducing the Temporal Difference family of iterative techniques to solve the Markov Decision Process
Image by Alexey Savchenko on Unsplash

In the previous article — Part 2 — we discovered a few solution algorithms to solve the Markov Decision Process (MDP), namely the Dynamic Programming method and the Monte Carlo method. The Dynamic Programming approach can be easily applied when we know the entire environmental dynamics of the MDP, such as the Transitional Probabilities between all states (conditioned on actions). However, such assumptions may not be practical, especially when we consider real-world applications, when stochastic relationships between states and actions are often vague.

Without knowledge of Transitional Probabilities, we then introduced the experiential learning with the… Read the full blog for free on Medium.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI