| (*)Students possess a solid understanding of numerical optimization techniques and their applications, learning to solve both unconstrained and constrained optimization problems. They are able to implement and apply numerical methods, such as gradient descent, Newton's method, and conjugate gradients, and understand the theory behind convex optimization and its significance in AI.
|
(*)- Applying Basic Optimization Principles (k4)
Students can understand and apply fundamental optimization principles, formulating and analyzing optimization problems across different contexts.
- Solving Equations with Numerical Methods (k4)
Students are able to use numerical methods for solving equations, including iterative techniques, ensuring the accurate and efficient solution of non-linear equations.
- Implementing Unconstrained Optimization Techniques (k4)
Students can apply optimization algorithms like Cauchy’s method (steepest/gradient descent), Newton's method, and the conjugate gradient method to find optimal solutions for unconstrained problems.
- Understanding and Solving Constrained Optimization Problems (k4)
Students are capable of tackling constrained optimization challenges, understanding how constraints affect solutions and utilizing techniques to solve such problems effectively.
- Applying Convex Optimization Methods (k4)
Students can identify and solve convex optimization problems, particularly linear and quadratic optimization, understanding their properties and efficient solution techniques.
- Analyzing and Comparing Optimization Methods (k5)
Students are able to analyze and compare the convergence, efficiency, and applicability of different optimization techniques to select the most appropriate method for specific AI tasks.
|
(*)Students have acquired foundational knowledge in optimization theory, covering key topics like unconstrained and constrained optimization, numerical methods for solving equations, and convex optimization principles. They understand how to implement and analyze various optimization algorithms, such as gradient descent, Newton's method, conjugate gradients, and techniques for solving linear and quadratic optimization problems relevant to AI applications.
|
