Learn the foundations of Machine Learning and build neural-network applications in Java with Deep Netts.
This course is intended for experienced Java developers who are new to Machine Learning. It introduces the concepts first, then connects them to compact Java examples that demonstrate the complete model-development workflow.
Complete Prerequisites, Download, Installation, and Hello World. Hello World confirms that Java, Maven, Deep Netts, licensing, and the Vector API work together.
The examples are standard Maven projects and are not tied to a particular IDE. Run them from a terminal or any Maven-compatible Java IDE using Java 25, and make sure --add-modules=jdk.incubator.vector is passed to the Java process that starts the application.
The lessons explain features, targets, Linear and Logistic Regression, neural-network architecture, backpropagation, Convolutional Neural Networks, the Deep Netts API, the complete model-development workflow, Natural Language Processing, and how those foundations connect to Generative AI.
Each Example page explains the problem, dataset, model, and important code before linking to a standalone Maven project:
Use the Glossary whenever you encounter an unfamiliar Machine Learning or Deep Netts term.
Start with Machine Learning Basics, Linear Regression, and Logistic Regression; continue with Neural Networks, Backpropagation, Convolutional Neural Networks, the Deep Netts API, Model Development, and Natural Language Processing; then finish with Generative AI.
The lessons explain concepts and compact code fragments. Use the Examples section for complete standalone Maven projects.
Begin with regression to understand inputs, targets, training, and error. Then continue with binary and multiclass classification.
Use the Glossary for concise definitions and return to the relevant lesson for the full context.
