Author(s): Rem E

Originally published on Towards AI.

Tutorial 9.1: Implementing the SARSA Algorithm for Our Maze Problem

Now we’re ready to start implementing our first Temporal Difference (TD) method: SARSA!
This tutorial builds on Tutorial 8.2, so make sure to check that one out first if you haven’t already.
If you’re not yet familiar with the SARSA algorithm, take a look at my previous story: Temporal Difference Learning: The Most Powerful RL Solution.
Ready? Let’s dive into the implementation!

This tutorial covers the implementation of the SARSA algorithm, explaining its core components and how it improves learning through the utilization of the Q-value updates for selecting actions and refining the policy over time. It details the project setup, initialization of variables, including the learning rate and discount factor, as well as the epsilon decay strategy to optimize action selection leading to the discovery of near-optimal policies.

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