AI Vocab Terms - Flashcards

Teaching computers to learn from examples and improve over time without being explicitly programmed for every scenario

A computer system modeled after the human brain that processes information through interconnected nodes to recognize patterns

A type of machine learning that uses multiple layers of artificial neurons to learn increasingly complex patterns from data

Technology that helps computers understand, interpret, and generate human language like we speak and write

The ability of computers to identify and understand objects, people, and scenes in images and videos like humans do

A collection of examples used to teach an AI system what to look for and how to respond

A step-by-step recipe or set of rules that a computer follows to solve a problem or complete a task

Teaching AI by showing it labeled examples where the correct answers are already provided

Letting AI find patterns and groupings in data on its own without being told what the right answers are

Training AI through trial and error by rewarding good decisions and penalizing bad ones, like training a pet

When an AI memorizes training examples too well and can't adapt to new, slightly different situations

The final trained AI system that has learned patterns from data and can now make predictions or decisions

When an AI system makes unfair or skewed decisions because the data it learned from contained prejudices or wasn't representative

A computer program that can have text or voice conversations with people to answer questions or help with tasks

AI that can create new content like text, images, music, or videos rather than just analyzing existing information

An AI system trained on massive amounts of text that can understand and generate human-like written content

A small chunk of text (like a word or part of a word) that AI breaks language into to process it

Taking an already trained AI and teaching it specialized skills or knowledge for a specific purpose

The instructions or questions you give to an AI system to tell it what you want it to do

When a trained AI uses what it learned to make predictions or generate responses to new inputs

Reusing knowledge an AI gained from one task to help it learn a different but related task faster

A setting or knob that humans adjust before training to control how an AI learns

How often the AI gets the right answer or makes correct predictions

A measurable characteristic or attribute of data that AI uses to make decisions or predictions

When AI sorts things into categories, like identifying whether an email is spam or not spam

When AI predicts a number or continuous value, like estimating house prices or tomorrow's temperature

When AI automatically groups similar things together without being told what the categories should be

AI determining whether text expresses positive, negative, or neutral emotions or opinions

AI identifying and locating specific things within images or video, like finding all the cars in a photo

A modern AI architecture that's especially good at understanding relationships between words in text

A hypothetical AI that could understand and learn any intellectual task a human can do, not just specialized tasks

The period when an AI is learning from examples and adjusting itself to get better at its task

An internal value that the AI adjusts during learning to improve its performance, like tiny dials being tuned

Creating modified copies of training examples to give AI more varied data to learn from

A separate set of examples used to check how well the AI is learning during training without contaminating the test

A final set of examples never seen during training used to evaluate how well the AI performs on new data

AI systems designed so humans can understand why they made specific decisions rather than being a black box

Running AI directly on local devices like phones or cameras instead of sending data to remote cloud servers

A method AI uses to gradually adjust itself by making small improvements based on its mistakes

Training AI across many devices while keeping data private on each device rather than collecting it all centrally