/navigating_a_virtual_world_with_dynamic_programming

This is the code for "Navigating a Virtual World Using Dynamic Programming" by Siraj Raval on Youtube

Primary LanguagePythonBSD 2-Clause "Simplified" LicenseBSD-2-Clause

rl-intro

Reinforcement Learning Introduction

Coding Challenge - Due Date, Thursday December 30 2017, 12 PM PST \

This weeks coding challenge is to use Monte Carlo Prediction to help a bot navigate an RL world. Use this as a guide and use OpenAI's gym environment. Bonus points for good documentation, good luck!

Overview

This is the code for this video on Youtube by Siraj Raval as part of the "Introduction to AI for Video Games" series. We're going to compare and contrast policy and value iteration algorithms. Both are a type of dynamic programming.

Dependencies

Usage

Just run 'python frozen_lake.py' in terminal

Credits

Credits for this code go to root-ua. I've merely created a wrapper to get people started.