Machine Learning and Pattern Recognition

Language : [English][Bahasa Indonesia]
Berikut ini adalah semua artikel machine learning & pattern recognition. Artikel dibuat sebagai catatan pribadi dari kuliah Machine Learning Fall 2017 – NCTU yang saat ini saya ikuti. Adapun textbook yang digunakan adalah “Pattern Recognition & Machine Learning” oleh Christopher M. Bishop. Artikel sudah diurutkan sesuai dengan alur materi.

• Regression problem
  1. Regresi Linear Menggunakan Least Square Estimation
  2. Polynomial Regression Menggunakan Least Square Estimation
  3. Menggunakan Regularization di Polynomial Regression untuk Mencegah Overfitting
• Optimization
  1. Taylor Series, Cara Kerja dan Penurunan Newton’s Method untuk Optimisasi
  2. Cara Kerja Optimisasi Gradient Descent untuk Menurunkan Eror
• Information Theory
  1. Beberapa Konsep Penting dalam Teori Informasi yang Berguna dalam Machine Learning
• Machine Learning dari sudut pandang probability (hampir semuanya membahas tentang Bayesian method)
  1. Pengenalan tentang Probability dan Random Variable
  2. Conditional Probability, Bayesian Formula dan Penurunan Naive Bayes Classifier
  3. Macam-macam Probability Distribution yang Penting untuk Memahami Online Learning
  4. Memahami Online leaning/sequential Learning pada Bayesian Inference
  5. Menurunkan Rumus Distribusi Gaussian
  6. Menggunakan Distribusi Gaussian untuk Online Learning/Sequential Learning
  7. Memahami Multivariate Gaussian (Gaussian dengan Multivariabel), Gaussian Properties dan Gaussian Mixture Model
  8. Bayesian Regression #Part1 : LSE vs Bayesian Regression, dan Menurunkan Posterior Update
  9. Bayesian Regression#Part2 : Menurunkan Predictive Distribution
• Estimator
  1. Estimator yang Bagus untuk Menghitung Mean dan Variance dari Sampel Data
• Klasifikasi
  1. Pengenalan tentang Klasifikasi dan Confusion Matrix
  2. Cara Kerja k-NN (k-Nearest Neighbors) untuk Klasifikasi
  3. Bias-Variance Decomposition
  4. Cara Kerja Logistic Regression untuk Binary Classification
  5. Memahami Metode Kernal dalam Machine Learning
• Klastering
  1. Klastering Menggunakan EM (Expectation Maximization) Mixture Model

Language : [English][Bahasa Indonesia]
Here are all machine learning and pattern recognition articles. These articles are made for my personal lecture notes in Machine Learning Fall 2017 – NCTU course I am joining in this semester. For the textbook, it uses “Pattern Recognition & Machine Learning” by Christopher M. Bishop. The article order is made regarding the material flow (up to mid term only).

• Regression problem
  1. Linear Regression using Least Square Estimation
  2. Polynomial Regression Using Least Square Estimation
  3. Deriving Polynomial Regression with Regularization to Avoid Overfitting
• Optimization
  1. Taylor Series, How It Works till Deriving Newton’s Method for Optimization
  2. How Gradient Descent Optimization Works for Minimizing Error
• Information Theory
  1. Some Important Information Theory Concepts in Machine Learning
• Machine learning from probability point of view (almost all talking about Bayesian method)
  1. Introduction to Probability and Random Variable
  2. Conditional Probability, Bayes’ Formula till Deriving Naive Bayes Classifier
  3. Some Important Probability Distributions to Understand Online Learning in Bayesian Inference
  4. Understanding Online/sequential learning in Bayesian Inference
  5. Deriving Gaussian Distribution
  6. Using Gaussian Distribution for Online Learning/Sequential Learning in Bayesian inference
  7. Understanding Multivariate Gaussian, Gaussian Properties and Gaussian Mixture Model
  8. Bayesian Regression #Part1: Prove LSE vs Bayesian Regression and Derive Posterior Update
  9. Bayesian Regression#Part2 : Deriving Predictive Distribution
• Estimator
  1. Good estimator for Mean and Variance of Sampled Data
• Classification
  1. Introduction to Classification and Confusion Matrix
  2. How k-NN (k-Nearest Neighbors) Works for Classification
  3. Bias-Variance Decomposition
  4. How Logistic Regression Works for Classification (with Maximum Likelihood Estimation Derivation)
  5. Understanding Kernel Method/Trick in Machine Learning
• Clustering
  1. Clustering using EM (Expectation Maximization) Method Mixture Model