Skip to content
Premium Content

Convex Optimization and Duality - The Theory Behind Learning

Why some optimization problems are easy and others hard: convexity, Lagrange multipliers, KKT conditions, and duality — the backbone of machine learning

This chapter requires a subscription to access.

What you'll unlock:

  • 1. What Makes a Problem Easy? Convex Sets and Convex Functions
  • 2. Recognizing and Building Convex Functions
  • 3. Unconstrained Convex Optimization: Gradient Descent and Newton's Method
  • 4. Constrained Optimization and Lagrange Multipliers
  • 5. Inequality Constraints and the KKT Conditions
  • 6. Lagrangian Duality: The Shadow Problem
  • 7. Convex Optimization in Machine Learning: SVMs, Regularization, and Beyond
Subscribe to Unlock

Already have an account? Sign in