AI Campus • Practical resources
AI & Neural Fundamentals
Learn what a model predicts, how training differs from inference and why confident output can still be wrong.
Before you start
- Basic algebra and ability to read simple charts. Python helps with the optional exercises.
The reading and planning exercises need no paid course or API. Optional inference, training, IDE features and hosted services may require hardware, accounts or paid access. Use the local or manual alternative before spending money.
Start with the official learning material
Free-to-read references checked 3 November 2026. Reading documentation does not activate a provider or create a subscription.
Practice with a concrete exercise
1. Learn the prediction loop
Read the regression, classification and neural-network sections. Draw a diagram separating training data, learned parameters, inference input and predicted output. Mark where an evaluation set enters the process.
2. Work a tiny example
Create ten fictional study-session rows with minutes studied and quiz score. Reserve three rows for testing. Explain why evaluating on the same seven training rows can overstate performance; inspect an outlier rather than deleting it silently.
3. Explain the limits
Write a one-page glossary covering token, embedding, parameter, context window and inference. For each, give a concrete example and one misconception. Describe one unsupported claim a language model could produce.
Check what you learned
- Your diagram keeps test examples separate from training.
- You can explain why prediction confidence is not factual proof.
- You identify which parts of the exercise are conceptual rather than a trained neural network.
Keep your artifact, observed checks and remaining limitations in a short README. Mark any exercise you only designed rather than executed.