AI glossary: terms starting with U

16 terms starting with U, each defined the way the lesson that teaches it uses it, with a link to every lesson where it appears.

underfitting

Fitting both poorly.

Taught in
underflow

a result closer to zero than the smallest float, stored as 0.0

Example: np.exp(-746)

Taught in
undocumented code

An observed source value without a category label in the declared map.

Taught in
Unicode

The standard for representing characters in Python strings; one indexed item is one code point and may be only part of a displayed character.

Taught in
uniform distribution

Equal-length intervals have equal probability: 2 to 4 minutes has 20% under the 0–10 model.

Taught in
union

Outcomes satisfying at least one of the events: 56,870 rows.

Taught in
Unit

one completed trip. The thing each row, and each count, is about.

Taught in
unit ball

every point with norm at most one

Example: circle, diamond, square

Taught in
unit of analysis

The thing counted once in a summary.

Taught in
unit vector

a vector of length one, a / ‖a‖

Example: (0.6, 0.8)

Taught in
univariate view

A view of one variable on its own, such as the distribution of daily invoice counts.

Taught in
Universal function

NumPy function applying an operation elementwise

Taught in
unobserved outcome

A result absent under the decision being evaluated, such as a rejected applicant's default. It is not a zero.

Taught in
unpacking

Binding each item of a sequence to its own name, in order: lat, lon = point.

Taught in
unsupervised learning

Finding structure in examples that have no labels.

Taught in
update rule

w ← w − η · L′(w).

Taught in