AI Basics

AI that could learn any new task, not just the one it was built for

A fixed sequence and set of rules for solving a problem

Often described as machines making people's judgment calls

The difference between following a fixed rule and judging as you go

Data piled up so high it forces you to change tools

A program you exchange text or speech with, back and forth

Picking one of the bins that were already decided in advance

Grouping similar things together with no labels at all

Technology that gets a computer to make sense of photos and video

The row-by-row record that becomes the raw material AI learns from

Turning scattered data into an answer worth acting on

A bundle of material gathered and organized to teach an AI

Machine learning that stacks processing steps many layers deep

Rewriting meaning as a set of numbers that marks a spot

One ingredient an AI looks at when it makes a judgment

The ability to keep performing well on data never seen before

AI that invents a result that never existed before

A device that processes the same calculation many times at once

The moment a trained model produces an answer for a new input

The answer paired with data for training, usually attached by a person

A language model grown enormous by reading a huge pile of text

A basic computing stage of neurons that share the same input

Finding rules from examples instead of writing them by hand

A block of rules saved as-is, left over once training finishes

AI built to do exactly one job well

Simple calculating units chained together in layers

A tiny computing unit that takes in many values and outputs one

The technology that lets computers handle human language

A tuning dial inside a model whose value training decides

Spotting the regularity that keeps repeating

Using what you know to guess at a value you don't

A system that picks out what you'll like from your past records

Naming a single number somewhere on a continuous scale

Learning by trying things and adjusting to the score you get back

The technology that transcribes spoken words into text

Learning a rule from examples that already carry the right answer

The bite-size pieces an AI cuts text into before reading it

The process of nudging dials a little at a time from examples

A test that guesses human or machine from conversation alone

Finding structure in data with no answers attached

The learned multiplier on each input, deciding how much it counts