MAT 581 Introduction to Numerical Methods
Fall 2020, Spring 2021, Fall 2021
Fall 2016
Exams
Unit 1: MATLAB
- Matrix/vector manipulation
- Matrix/vector creation; elementwise operations
- Built-in functions, input/output, 2D graphics
- Scripting
- Functions
- Linear systems
Unit 2: Root Finding
- Root finding: bisection method
- Fixed point method
- Newton and secant methods
- Multivariable Newton's method
- Broyden's method
Unit 3: Numerical Calculus
- Numerical differentiation
- Numerical integration: Trapezoidal and midpoint rules
- Simpson's rule, Newton-Cotes
- Orthogonal polynomials: Legendre, Laguerre
- Gaussian integration, Gauss-Laguerre integration
- Adaptive integration
- Multiple integrals
- ODE: Euler's method, RK2 methods
- ODE systems, predator-prey model
Unit 4: Data Fitting
- Polynomial interpolation
- Spline interpolation
- Spline approximation
- Discrete Fourier transform
- Linear least squares
- Model comparison
- Nonlinear least squares
- Transforming data
Unit 5: Optimization
- Linear programming
- Single-variable minimization
- Gradient method and conjugate gradient method
- Nelder-Mead method
- Applications of optimization: data classification
- Applications of optimization: model selection
Fall 2015
Homework (Matlab)
- Vectors and plotting
- Matrices
- Experiments with probability
- Floating point arithmetics
- Matlab functions
- Polynomial interpolation
- Newton polynomial
- Error of interpolating polynomial
- Flop Count
- Piecewise linear interpolation
- Adaptive piecewise linear interpolation
- Natural cubic splines, theory
- Construction of cubic splines
- B-splines
- Curve fitting by least squares
- More on least squares
- Training and testing data sets
- Root finding: bisection
- Root finding: Newton's method
- Root finding: secant method
- Root finding: Multivariable Newton's method
- Minimization: golden section search
- Minimization: gradient descent
- Multivariable minimization: gradient descent
- Multivariable minimization: Nelder-Mead
- Multivariable minimization: application to classification of data
- Numerical differentiation
- Numerical integration: midpoint and trapezoid
- Numerical integration: adaptive Simpson's rule
- Numerical integration: Newton-Cotes formulas
- Discrete Fourier transform
- Numerical integration: Clenshaw-Curtis method
- Discrete cosine transform, signal compression
- ODE: Euler's and Runge-Kutta methods
- ODE: boundary value problems