KodeKloud Notes

Artificial Intelligence (AI) is a field of computer science focused on creating software that mimics human intelligence. From Netflix recommendations and Amazon product suggestions to Gmail’s spam filters, AI operates behind the scenes to enhance user experiences.

Machine Learning (ML), a subset of AI, designs algorithms that learn patterns from historical data to make decisions or predictions. Deep Learning, a specialized branch of ML, leverages artificial neural networks to process large volumes of unstructured data such as images, audio, and text.

The image is an introduction to AI, illustrating the relationship between computer science, artificial intelligence, machine learning, and deep learning, with a focus on deep learning as a type of machine learning based on artificial neural networks.

Traditional Programming vs. Machine Learning

In traditional programming, you explicitly write rules (business logic) that transform input data into output. For example, converting US dollars to another currency involves fetching the current exchange rate and applying it.

Machine Learning flips this paradigm: it ingests historical exchange-rate data and learns the conversion pattern without you specifying the rule.

The image compares traditional coding and machine learning. Traditional coding uses data and rules to produce answers, while machine learning uses data and answers to derive patterns or rules.

Common Machine Learning Techniques

Machine Learning algorithms typically fall into three categories:

Technique	Data Type	Goal
Supervised Learning	Labeled data	Predict labels for new inputs
Unsupervised Learning	Unlabeled data	Discover patterns or clusters in the dataset
Semi-Supervised	Mixed data	Combine a small labeled set with large unlabeled data

The image is a diagram showing common techniques of machine learning, including supervised, unsupervised, and semi-supervised learning.

Supervised Learning

Supervised learning trains models on labeled datasets, teaching them to map inputs to known outputs. Once trained, the model generalizes to classify or predict labels on unseen data.

The image illustrates a supervised learning process with labeled images of a dog, cat, and rabbit being used to train a model.

Note

Ensure your labeled dataset covers diverse examples to avoid bias and improve model accuracy.

Unsupervised Learning

With no labels, unsupervised algorithms uncover inherent structures—such as clusters or hidden features—by analyzing the input data alone. This is useful for customer segmentation, anomaly detection, and feature extraction.

The image illustrates the process of unsupervised learning, showing unlabelled data of animal silhouettes being processed through training to create a model that predicts and categorizes the animals.

Semi-Supervised Learning

Semi-supervised learning bridges supervised and unsupervised methods by pairing a small amount of labeled data with a large pool of unlabeled data. This approach often yields high accuracy while reducing labeling costs.

The image illustrates semi-supervised learning, showing labeled and unlabeled animal icons and a model, with text explaining it as a mix of supervised and unsupervised learning requiring a smaller labeled dataset.

Warning

Quality of unlabeled data affects performance. Preprocess and clean data to avoid propagating errors.

Deep Learning

Deep Learning models consist of multiple layers of interconnected neurons, inspired by the human brain. These neural networks can learn complex representations across supervised, unsupervised, and semi-supervised tasks, making them highly versatile for image recognition, natural language processing, and more.

The image illustrates deep learning concepts, showing a brain, a neural network, and icons representing different learning types, with descriptions highlighting inspiration from the human brain and applications in various learning methods.

Summary

AI: Broad field that imbues machines with human-like intelligence.
Machine Learning: Subset of AI focused on learning patterns from data.
Deep Learning: Advanced ML using neural networks; excels with unstructured data and demands powerful hardware.

In the next lesson, we’ll dive into how these foundations power Generative AI models.

Links and References

Watch Video

Watch video content