Scribd
Upload
3 views4 pages

Supervised Learning

The document outlines three primary types of machine learning: supervised learning, where models learn from labeled data; unsupervised learning, which identifies patterns in unlabeled data; and reinforcement learning, where agents learn through interaction with their environment. It also mentions hybrid approaches like semi-supervised and self-supervised learning, which utilize both labeled and unlabeled data. Common applications include classification, regression, clustering, and recommendation systems.

Uploaded by

spark20645
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views4 pages

Supervised Learning

The document outlines three primary types of machine learning: supervised learning, where models learn from labeled data; unsupervised learning, which identifies patterns in unlabeled data; and reinforcement learning, where agents learn through interaction with their environment. It also mentions hybrid approaches like semi-supervised and self-supervised learning, which utilize both labeled and unlabeled data. Common applications include classification, regression, clustering, and recommendation systems.

Uploaded by

spark20645
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

TYPE 01

Supervised Learning
The most common approach — the model is trained on labelled examples, learning to map inputs to known outputs. Think of it
as a student learning from a teacher who provides the correct answers.

Labelled Model
Predict New
Data Learns

Common tasks include classification (spam detection) and regression (house price prediction). Popular algorithms: Decision
Trees, SVMs, Neural Networks.
TYPE 02

Unsupervised Learning
No labels, no teacher — the model discovers structure on its own. It
groups similar data points and reveals hidden patterns humans might miss.

Clustering Dimensionality
Groups similar data — e.g. Reduction
customer segmentation, Simplifies data — e.g. PCA,
anomaly detection visualising high-dimensional
datasets

Association
Finds relationships — e.g. market basket analysis, recommendation
engines
TYPE 03

Reinforcement
Learning
An agent learns by interacting with an
environment, receiving rewards for good
actions and penalties for bad ones. Over
time, it develops an optimal strategy —
much like training a dog with treats.

Used in robotics, game-playing AI


(AlphaGo), and autonomous vehicles.

Agent Environment Reward Signal


The learner or decision-maker The world the agent operates in Feedback guiding the learning
process
BEYOND THE BIG THREE

Semi-Supervised & Self-Supervised Learning


Not all learning fits neatly into supervised or unsupervised categories. These hybrid approaches bridge the gap, making efficient
use of limited labelled data.

Semi-Supervised Learning Self-Supervised Learning


Uses a small amount of labelled data alongside a large pool The model creates its own labels from the data structure.
of unlabelled data. Ideal when labelling is expensive — Powers modern language models (like GPT) and computer
common in speech recognition and web content classification. vision systems without human annotation.

You might also like