Gradient Descent: Logistic Regression

Explore how gradient descent is used to optimize logistic regression models for binary classification tasks, like predicting diabetes. Understand the role of the sigmoid function, loss minimization, and iterative parameter updates to find optimal solutions. Gain practical skills to implement this method with Python.

We'll cover the following...

Logistic regression
Gradient descent
Optimizing logistic regression

Logistic regression

Consider the scenario where we want to create a model that predicts whether an individual has diabetes or not. This is a case of binary classification and let’s assume that the input matrix $X = \begin{bmatrix} x_1^T \\ x_2^T \\ . \\. \\ . \\ x_N^T \end{bmatrix} \in \R^{N \times d}$ represents the collection of the $d$ -dimensional input features, such as age, weight, height, cholesterol, etc., for $N$ patients. $Y = [\begin{array}{c} y_{1} \\ y_{2} \\ . \\ . \\ . \\ y_{N} \end{array}] \in {0, 1}^{N}$ ...

1.Introduction to Optimization

2.Vector Calculus

3.Convex Optimization

4.Gradient Descent for Non-Convex Optimization

Project

5.Constrained Optimization

6.Miscellaneous Methods

7.Course Conclusion

Assessment

Mini Project

Gradient Descent: Logistic Regression

Logistic regression