Home › Technology › Artificial Intelligence & ML › Training

Training

TRAY-ning /ˈtreɪnɪŋ/

nounverbB1SATTOEFLIELTS

formalgeneraltechnical

(noun) The action of teaching a person or animal a particular skill or type of behavior. (in AI) The process of feeding data to a Machine Learning model so it can learn patterns and make predictions. (verb) To teach or educate.

In Machine Learning, 'training' involves iteratively adjusting the parameters of a model using a dataset (the 'training data') to minimize errors and improve its ability to perform a specific task. It's the learning phase.

The story behind the word. The word 'Training' comes from the Old French `trainer`, meaning 'to drag' or 'to draw.' This evolved in English to mean 'to educate' or 'to discipline,' often through practice or instruction, much like 'drawing out' skills from someone. In AI, it refers to the process of 'teaching' a model by repeatedly exposing it to data and adjusting its internal parameters until it performs its task effectively.

Word relationships

education: Broader, often more formal and academic.
instruction: The act of giving knowledge or skills.
coaching: More focused on personal development and specific skills.

Commonly confused with

inference: Training is the process of learning from data; inference is the process of applying that learned knowledge to new data.

Word family

train (verb): To teach or educate.
trainer (noun): A person who trains others.

Collocations

job training
staff training
intensive training
training data
model training

Idioms & expressions

in training

Example sentences

"The company provides extensive Training for new employees to ensure they are proficient in their roles." Discussing professional development.
"The quality of the Data used for Training is crucial for the performance and fairness of the AI Model." In a Machine Learning context.

Memory hook

Think of a personal 'trainer' who helps you learn new skills by practicing over and over. Practice to Learn

When not to use

Do not use 'training' to refer to the application of a learned model (which is 'inference'). Training is about the learning process itself.

Fun facts

The concept of 'training' a model is analogous to how humans learn from experience and examples.
The amount and quality of training data are often more important than the complexity of the algorithm for a model's performance.

Related words

Data, Model, Machine Learning

Explore more words

educate, conditioned, upskill, coaching, socialize, upbringing