Back to Courses
SCI1005 Undergraduate

WU | Intelligent Optimization

An advanced course covering the theory and application of intelligent optimization techniques, ranging from classical gradient methods to modern evolutionary meta-heuristics, multi-objective optimization, surrogate-assisted models, and federated machine learning applications.

5.0
45.0h
300 students
2 lessons
0 likes
Artificial Intelligence
Start Learning

Course Overview

📚 Content Summary

An advanced course exploring the intersection of classical mathematical optimization, evolutionary computation, and contemporary machine learning. It covers single and multi-objective optimization, data-driven strategies, and advanced topics like Neural Architecture Search and Federated Learning.

Master the evolution of intelligence through advanced data-driven and multi-objective optimization.

🎯 Learning Objectives

  1. [Apply advanced optimization techniques to complex, data-driven mathematical problems.]
  2. [Design and evaluate evolutionary and machine learning architectures for multi-objective scenarios.]

Lessons 共 2 课时 · 预计 45.0h

Lessons

Lesson

This lecture introduces the fundamentals of optimization by defining core components such as decision vectors, objective functions, and constraints within a search space. Students will learn to distinguish between classical gradient-based methods, which are effective for convex problems, and stochastic evolutionary algorithms designed to navigate complex, non-convex landscapes.

This lesson introduces Genetic Algorithms as stochastic search heuristics, focusing on the distinction between genotypes and phenotypes and the importance of effective variable representation. Students learn to implement the evolutionary loop—including selection, crossover, and mutation—while addressing technical challenges like Hamming cliffs through binary, Gray, and real-coded mapping techniques.