Vocabulary

Understand the term. Connect the concept.

A practical dictionary of 512 AI, machine learning, mathematics, and programming terms from HerbDev’s AI Vocabulary app. Start with a definition, explore an example, then follow the concepts behind it.

Linked concepts open in a new tab. Audio reads the definition and example.

Similar words, different decisions.

Compare the definitions, then use the example to see why the distinction matters.

Parameter vs. Hyperparameter

Parameter

A named input in a function definition, or an adjustable value in a model.

Hyperparameter

A setting chosen outside the ordinary parameter-fitting process.

What the model learns versus what you configure

In model training, a parameter such as a weight is fitted from data. A hyperparameter such as the learning rate controls how training runs. In programming, parameter also means a named function input; the context matters.

Example

Training can update a network’s weights while you hold its learning rate fixed. Hyperparameter search can optimize the learning rate in a separate process.

Accuracy vs. Precision

Accuracy

The fraction of evaluated predictions that are correct.

Precision

The fraction of predicted positives that are truly positive.

All predictions versus predicted positives

Accuracy divides correct predictions by all evaluated predictions. Precision divides true positives by predicted positives. They answer different questions, especially when classes are imbalanced.

Example

Suppose 10 of 1,000 cases are positive. Predicting negative for every case gives 99% accuracy but finds none of the positives; positive-class precision is undefined because there are no positive predictions. Pair these measures with recall and the actual costs of errors.

Embedding vs. Memory

Embedding

A numerical representation placing items in a vector space.

Memory

Stored information made available to later agent steps or sessions.

Representation versus retained information

An embedding represents an item numerically. Memory makes past information available later. An embedding can be stored as part of memory, but creating one does not automatically retain it. This dictionary’s memory entry concerns agent information, rather than hardware RAM or a specific persistent model-state mechanism.

Example

Encode a photograph and discard the vector: you created a representation. Save that vector with its source and capture date: it becomes part of a retrievable history.

Manage browser learning data

“Remove my test data” clears all AI 101 and ML 101 scores, reviews, project checklists and rating selections, plus saved vocabulary and recent history in this browser. Test entries cannot be separated from other learning progress. Cookie preferences, security protections, submitted feedback, contact emails and past analytics are kept.