Learning
Separate learned parameters from information supplied with a request.
Skip to content
AuraUnderstand AI
Understand AI
Choose your starting point
Learn a concept, test it, then examine its foundations. Every level is directly accessible.
I’m getting started
Distinguish learning from inference before exploring mechanisms.
Start with the basicsI want to go deeper
Equations, assumptions, differences between approaches and original papers.
Go directly to advanced content01First principles
Inference
Separate the model’s probabilities from the rule that selects an output.
02Mechanisms
Representations
Understand why words become vectors without equating a vector with human meaning.
Context & memory
Separate working context, recurrent state, learned parameters and persistent storage.
Architectures
Compare architectures, training methods and complete systems at the right level.
Diffusion
Distinguish corruption during learning from progressive generation.
03Research & evaluation
Reasoning & agents
Understand the harness around a model and its tools.
World models
Explore longer horizons and the risk of inaccurate predictions.
AGI & ASI
Examine generality, performance and adaptation before applying a label.
Evaluation
Move from a convincing impression to a reproducible comparison.