🎬AI Lectures28

This session sets up course logistics and introduces core machine learning ideas. You learn when and how class meets, where to find materials, how grading works, and why MATLAB is used. It also sets expectations: the course is challenging, homeworks are crucial, and live attendance is encouraged.

This lecture introduces supervised learning for regression, where the goal is to predict a real number (like house price) from input features (like square footage, bedrooms, and location). You represent each example as a d-dimensional vector x with a target y. Linear regression models this relationship with a straight-line formula: f(x) = w^T x + b. The focus is on learning weights w and bias b that best map inputs to outputs.

The lecture explains why we use machine learning instead of writing step-by-step rules. Many real problems, like finding cats in photos, are too messy for hand-written rules because there are too many exceptions. With machine learning, we give the computer lots of examples and it discovers patterns on its own. This approach lets computers handle tasks we can’t easily explain in code.

Decision trees are models that make predictions by asking a series of yes/no questions about features, like a flowchart. You start at a root question, follow branches based on answers, and end at a leaf that gives the prediction. This simple structure makes them easy to read and explain to anyone.

Logistic regression is a simple method for binary classification that outputs a probability between 0 and 1 for class 1. It takes a weighted sum of input features (w^T x + b) and passes it through the sigmoid function. The sigmoid is an S-shaped curve that squashes any real number into the [0,1] range.

Machine learning lets computers learn patterns from data instead of following hand-made rules. Instead of writing instructions like “pointy ears = cat,” we feed many labeled examples and let the computer discover what features matter. This makes ML flexible and powerful for messy, real-world problems where rules are hard to write. Arthur Samuel’s classic definition captures this: computers learn without being explicitly programmed.

This lecture explains L1 regularization, also called LASSO, as a way to prevent overfitting by adding a penalty to the loss that depends on the absolute values of model weights. Overfitting means a model memorizes the training data but fails on new data. By penalizing large weights, L1 helps the model focus on the strongest, most useful features.

Decision trees are simple, powerful models that make predictions by asking a sequence of yes/no questions about input features. They work for both classification (like spam vs. not spam) and regression (like house price prediction). The tree’s structure—what to split on and when—comes directly from the data, which is why we call them non-parametric.

The lecture explains what deep learning is and why it changed how we build intelligent systems. In the past, engineers wrote step-by-step rules (like detecting corners and lines) to identify objects in images. These hand-built rules often broke when lighting, angle, or season changed. Deep learning replaces these hand-crafted rules with models that learn directly from data.

Artificial Intelligence (AI) is the science of making machines do tasks that would need intelligence if a person did them. Today’s AI mostly focuses on specific tasks like recognizing faces or recommending products, which is called narrow AI. A future goal is general AI, which would do any thinking task a human can, but it does not exist yet.