AI glossary: terms starting with P

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

p-hat

Estimate of p from these observed rows: 495 / 3,150.

Taught in
package

A folder of modules, imported with dotted names: review_stats.text_tools.

Taught in
packing

Collecting several values, separated by commas, into one tuple: point = 12.9716, 77.5946.

Taught in
pair plot

A grid of pairwise plots with one-variable distributions on the diagonal.

Taught in
pair plot, heatmap

A scatter for every pair of features; a grid of cells coloured by each pair's correlation.

Taught in
paired comparison

Two results measured on the same observational units.

Taught in
paired resampling

Applying one draw of unit indices to every measurement on those units.

Taught in
palindrome

Text that reads the same forwards and backwards.

Taught in
parameter

The name in a function's definition that a value will be bound to: price in def with_tax(price):.

Taught in
parameter p

The assumed chance of outcome one on one modelled trial; here set to 15.71%.

Taught in
partial derivative

the derivative with respect to one variable, the others held fixed

Example: ∂/∂w (y − wx − b)² = −9.6 for the 4-year-old car

Taught in
path

The folders and the file name that locate a file: Path("data") / "notes.txt".

Taught in
Pearson correlation

Standardized linear association between two numerical variables: covariance divided by the two standard deviations. Written r, between -1 and 1.

Taught in
penalty

a norm of the weights added to a model's error

Example: ridge: ‖w‖₂²; lasso: ‖w‖₁

Taught in
penalty strength

The hyperparameter that sets how much the penalty counts: alpha in scikit-learn, λ in papers.

Taught in
percentile

A quantile whose fraction is written as a percentage: the 90th percentile is the 0.9 quantile.

Taught in
percentile bootstrap interval

Bounds read from specified quantiles of a bootstrap distribution.

Taught in
perfect collinearity

A feature that is an exact linear function of other features (singular matrices: Maths 3.3, second pass).

Taught in
Permutation

Reordering of positions without duplication

Taught in
pipeline

Preprocessing and a model fitted together, so each step learns from the fitting rows only.

Taught in
plane

w · x + b = 0 in three dimensions

Example: x + 2y + 2z − 6 = 0

Taught in
point cloud

a table drawn as one point per row

Example: the ward's 14 patients

Taught in
point estimate

One statistic computed from one sample.

Taught in
polynomial features

The powers added as columns.

Taught in
polynomial regression

Linear regression on powers of a feature.

Taught in
Population

the trips the decision is meant to cover. In 00.2, completed yellow-taxi trips picked up in Midtown Center on weekdays from 09:00 to 09:59, now and under the same conditions later.

Taught in
population change

A changed set of cases in a summary.

Taught in
population mean

Average over every unit in the declared frame: £534.40.

Taught in
positional argument

An argument matched to a parameter by its position.

Taught in
positive skew

A distribution with a longer or heavier right tail.

Taught in
posterior

Probability after incorporating the stated evidence: 8.33% after one flag.

Taught in
posterior distribution

Uncertainty over the rate after updating the declared prior with observed labels.

Taught in
posterior predictive probability

The model's chance of one next outcome after averaging over posterior rate uncertainty.

Taught in
potential confounder

A third variable associated with predictor and outcome that could alter their interpretation.

Taught in
power

bⁿ

Example: 2⁵ = 32

Taught in
power rule

the derivative of xⁿ is n xⁿ⁻¹

Example: x³ → 3x², 12 at 2

Taught in
precedence

The order operators run in: `, then unary -, then * / // %, then + -, then comparisons, then not, and, or`. Round brackets override it.

Taught in
predict

The method that gives answers for new inputs, using what fit learned.

Taught in
prediction time

The moment the model is used: for the bank, before anyone dials.

Taught in
predictive value

Measured improvement on evaluation data that did not guide the exploration.

Taught in
predictor

Recorded input used to calculate a fitted value; here, trip distance.

Taught in
prior

Probability before the new evidence: 1% spam in the constructed inbox.

Taught in
prior distribution

Assigns probability across the possible rates before the reviewed rows supply any evidence.

Taught in
prior sensitivity

A change in posterior results caused by changing the assumed prior while keeping evidence fixed.

Taught in
probabilistic label

The fraction of the K neighbors with each label.

Taught in
probability density function, PDF

Curve whose interval areas give probabilities.

Taught in
probability ellipse

The region inside a squared Mahalanobis radius that holds a stated share of the model's probability.

Taught in
product notation Π

multiply an expression over an index

Example: Πᵢ₌₁³ i = 6

Taught in
product rule

log(ab) = log a + log b

Example: ln 0.1 = −2.3026

Taught in
program

A set of instructions for a computer, written as text.

Taught in
programming language

Precise rules for writing instructions a computer can carry out.

Taught in
project

A folder of code with a pyproject.toml that describes it.

Taught in
projection

the part of a vector along a direction, (b · â) â

Example: (1.2, 1.6)

Taught in
pytest

A tool that finds the test functions in a project and runs them all.

Taught in