Articles → MACHINE LEARNING → Models In Machine Learning

Models In Machine Learning

This article describes Models in Machine Learning.

Mathematical Representation

A mathematical representation is a way of writing the set of rules or relationships using numbers and symbols instead of sentences.

Sentence

The cost of apples is 10 rupees each. If you buy 3 apples, you pay 30 rupees.

Mathematical Representation

Cost = 10 × Number of Apples

What Is A Model?

A model is the mathematical representation of the real-world process that machine learning uses to make predictions and decisions.

Inputs, Outputs, And Parameters

Inputs

These are the values you feed into the model.

Outputs

These are the values you want the model to predict.

Parameters

Parameters are internal knobs that the model adjusts so it can map input → output correctly.

Parameter = Model Memory

A model’s parameters store what it has learned from the data. When the parameters are updated (during training), it means the model is adjusting its internal knobs to better map inputs → outputs.

Constituents Of A Machine Learning Model

A model usually consists of the following components.

Component	Description
Parameters	learned values
Hyperparameters	manually set values that control learning
Loss Function	a metric to measure error
Optimizer	algorithm to update parameters
Model Architecture	structure (e.g., linear, decision tree, neural net).
Training Data	The input-output examples that teach the model.
Prediction Function	how the model transforms input into output.

Posted By -	Karan Gupta

Posted On -	Thursday, September 25, 2025

Query/Feedback

Your Email Id		**

Subject		*

Query/Feedback	Characters remaining 250	**