Skip to main content

Ctrl+K

FMMS050 - Statistical regression and machine learning

PART 1 -- MACHINE LEARNING FOR ENGINEERING APPLICATIONS

Types of machine learning methods
Basic mathematics for machine learning
Lecture 1 -- AI/ML terminologies

PART 2 -- STATISTICAL REGRESSION

Lecture 2 -- Regression and interpretation
- Computer Examples
- Computer Excercises
Lecture 3 -- Polynominal and spline fitting
- Computer Examples
- Computer Excercises
Lecture 4 -- Model parameter estimation by gradient
- Computer Examples
- Computer Excercises
Lecture 5 -- GLM GAM and Mixed-effects model
- Computer Examples
- Computer Excercises

PART 3 -- MACHINE LEARNING METHODS

Lecture 6 -- Logistical regression and classification
- Computer Examples
- Computer Excercises
Lecture 7 -- Decision trees and ensemble algorithm
- Computer Examples
- Computer Excercises
Lecture 8 -- Boosting methods (XGBoost)
- Computer Examples
- Computer Excercises
Lecture 9 -- Support vector machine
Lecture 10 -- Artificial neural network

PART 4 -- TIME SERIES LEARNING METHODS

Lecture 11 -- Gaussian transformation methods
- Computer Examples
- Computer Excercises
Lecture 12 -- Basic properties of stationary Gaussian process
Lecture 13 -- Time series analysis and model exploration
Lecture 14 -- Auto regressive models and Moving average models (1)
- Computer Examples
Lecture 15 -- Auto regressive models and Moving average models (2)
Lecture 16 -- ARIMA models

PROJECT ASSIGNMENT

Project 1 -- Simple regression for roll damping
Project 2 -- Machine learning models for performance modelling
Project 3 -- ARIMA model for ice thickness prediction
Students self defined projects

PART 5 -- DATA FILTERING METHODS (upcoming)

Lecture 17 -- Fast Fourier Transform
Lecture 18 -- Moving Average Methods
Lecture 19 -- Kalmar filtering
- Tutorial of Kalmar filter methods

PART 6 -- DEEP LEARNING AND SPATIO-TEMPORAL MODELLING (upcoming)

Lecture 20 -- Convolutional neural network
Lecture 21 -- Recursive neural networks - LSTM
Lecture 22 -- Spatio-temporal modelling
Lecture 23 -- Statistical paritical differential equations

Repository
Suggest edit
Open issue

.md

Moving average

Moving average#

Moving average method is an important tool to smooth data and clean unexpected/unrealistic spikes from the original measurement data.

previous

Fast Fourier Transform

next

Kalmar filter

By Wengang Mao

© Copyright 2023-2024.