[AI / ML] Supervised Learning
For dummies… like myself.
Complicated Definition
Stuff and Junk
Vocab:
A class of problem
a classification that involves predicting class labels and regression that involves predicting a numerical value
Classification: Supervised learning problem that involves predicting a class label
Regression: Supervised learning problem that involves predicting a numerical label.
a mapping
input example
target variable
input vector
target vector
models are fit
decision trees
supper vector machines
A feature vector
A vector
matrix
tree
graph
In supervised learning, the dataset is the collection of labeled examples
$$ \{(x_i, y_i)\}_{i=1}^{n} $$
Each element of x, among N is called a feature vector.
a feature vector is a vector in which each dimension j = 1, …., D contains a value that describes the example somehow.
That value is called a feature and is denoted as x^(j)
if each example x in our collection represents a person, then the first feature, x^(1), could contain height in cm
Supervised algorithms are called such because they “learn” by making predictions given examples of input data, and the models are supervised and corrected via an algorithm to better predict the expected target outputs in the training dataset.
The term supervised learning originates from the view of the target Y being provided by an instructor or teacher who shows the machine learning system what to do.
Some algorithms may be specifically designed for classification, such as logistic regression or regression such as linear regression and some may be used for both types of problems with minor modifications such as artificial neural networks.
The goal of a supervised learning algorithm is to use the dataset to produce a model that takes a feature vector X as input and outputs information that allows deducing the label for this feature vector.
For example: the model created uses as dataset of people that has an input vector which describes a person and outputs the probability that person has cancer based on the j features in the feature vector.
Fantasy / Sci-Fi Example
Supervised Learning in Machine Learning: A Magical Guide for Beginners
Have you ever wished you had a magical spell to predict the future? Well, in the world of Machine Learning (ML), we have something close: Supervised Learning! Think of it as a wise wizard training their apprentice, teaching them to recognize patterns in the mystical world around them.
What is Supervised Learning?
Supervised Learning is a type of ML where we train a model using labeled data. Imagine a great magical archive filled with ancient spellbooks, each containing spells (input data) and their effects (output labels). Our goal is to teach a young sorcerer (our ML model) to understand these connections so they can predict the outcome of any new spell they encounter.
How Does It Work?
Gather the Data (The Magic Library) – We collect a dataset with inputs and correct outputs. For example, a kingdom’s alchemy guild keeps records of magical ingredients and the potions they create.
Train the Model (The Sorcerer’s Training) – Our model studies the dataset, learning which ingredient combinations (inputs) lead to specific potion effects (outputs).
Make Predictions (Casting Spells) – Once trained, the model can predict the effect of a potion it has never seen before based on past knowledge.
Evaluate Performance (Spell Accuracy Tests) – We test the model on new data to see how well it predicts outcomes. If the model misfires too often, it needs more training!
Example: Predicting Magical Creatures
Let’s say we live in a world where different magical creatures exist—dragons, griffins, and unicorns. Each creature has unique characteristics:
Dragons breathe fire, have scales, and live in volcanoes.
Griffins have wings, lion bodies, and prefer high altitudes.
Unicorns have horns, eat enchanted herbs, and reside in sacred forests.
We create a dataset with features (size, habitat, special abilities) and labels (creature type). Our model learns from this data and, given a new set of features, can predict whether the creature described is a dragon, a griffin, or a unicorn.
Types of Supervised Learning
Classification (Sorting Magic) – The model predicts categories. Example: "Is this creature a dragon, a griffin, or a unicorn?"
Regression (Predicting Spell Potency) – The model predicts continuous values. Example: "How powerful will this potion be on a scale of 1-100?"
Real-World Fantasy Uses
A wizard’s crystal ball that predicts a kingdom’s future prosperity based on trade, weather, and dragon activity.
A magical archivist that sorts lost scrolls based on the kind of spells they contain.
A seer’s enchanted mirror that estimates how long a hero will survive in a dungeon based on their armor, weapons, and experience.
Final Thoughts
Supervised Learning is like a wise master training an apprentice. With the right data and practice, the apprentice (our model) becomes a powerful predictor of the unknown! Whether you’re sorting magical creatures, brewing potions, or deciphering ancient scrolls, this ML method is your gateway to predictive magic.
So, what will you train your magical model to predict? Let me know in the comments!
