A Raspberry Pi and DeepLearning4J Brain That Plays Minecraft

Explore how to create an autonomous Minecraft-playing system using a Raspberry Pi cluster and DeepLearning4J in this 50-minute Devoxx conference talk. Dive into the fundamentals of machine learning, neural networks, and Markov decision processes before examining the architecture of a Raspberry Pi cluster. Learn about the DeepLearning4J library and its application in Java development for AI. Discover how Project Malmo, an AI experimentation platform built on Minecraft, integrates with these technologies. Gain insights into the software architecture designed for autonomous Minecraft gameplay and witness a live demonstration of the Pi Brain in action. Conclude with a discussion on future directions and potential applications of this innovative approach to AI in gaming.

Intro
Inspiration
Hierarchical Temporal Memory
Introduction To Machine Learning
Deep Learning Basics
Neural Network Elements: Node
Neural Network Elements: Layers
Markov Decision Process
Raspberry Pi Variations
Pi Brain Architecture
Raspberry Pi Cluster (Mk 2)
Cluster Hat
Why Java?
Azul Zulu Java For Embedded
Deep Learning4J (DL4J)
DL4J Core Classes
RL4J Basics
Project Malmo
Malmo Example Code (1)
Demo Software Architecture
Applying ML To Minecraft
Conclusions
Future Directions