ml-study-notes.md

ML study notes

모두를 위한 머신러닝/딥러닝 강의

• Lecture
• Code

Lecture 1: Machine Learning Basics

• What is ML?
• What is learning?
  • supervised
  • unsupervised
• What is regression?
• What is classification?

Machine Learning

• explicit programming(ex. a일 때는 b를 하라) 의 한계
• ML: "Field of study that gives computers the ability to learn without being explicitly programmed" by Arthur Samuel

Supervised/Unsupervised learning

• Supervised learning:
  • 레이블이 정해져 있는 데이터(training set)으로 학습
• Unsupervised learning:
  • un-labeled data
  • ex. 구글 뉴스 그룹핑, 단어의 분류 Word clustering

Supervised Learning

• Most common problem type in ML
• 이미 트레이닝 셋이 있음, 새로운 셋을 트레이닝 데이타 셋을 기반으로 분류

Types of supervised learning

1. Predicting final exam score based on time spent => regression
2. Pass/non-pass based on time spent => binary classification
3. Letter grade (A, B, C, D and F) based on time spent => multi-label classification

Lecture 2: Linear Regression

• 이미 존재하는 데이터(training data set)로 Regression model을 학습시킴(training)
• 세상의 많은 현상들이 리니어한 모델로 설명할 수 있다!
• (Linear) Hypothesis: H(x) = Wx+b
• Which hypothesis is better? => Cost Function(Loss Function)
• cost(W, b) = average of (H(x)-y)^2
• Goal: Minimize cost(W, b)

Lecture 3: How to minimize cost

Simplified Hypothesis

Gradient descent algorithm

• 경사를 따라 내려가는 알고리즘
• Minimize cost function
• Gradient descent is used many minimization problems
• For a given cost function, cost(W, b), it will find W, b to minimize cost
• It can be applied to more general function: cost(W1, W2, ...)

How it works? How would you find thd lowest point?

• Start with initial guesses
  • Start at (0,0) (or any other value) 아무 점에서 시작한다.
  • Keeping changing W and b a little bit to try and reduce cost(W, b)
• Each time you change the parameters, you select the gradient which reduces cost(W, b) the most possible
• Repeat
• Do so until you converge to a local minimum
• Has an interesting property
  • Where you start can determine which minimum you end up
• 기울기는 미분해서 구한다. 기울기 < 0 이면 W를 크게, 기울기 > 0 이면 W를 작게 만든다.
• cost(W, b)를 설계할 때 그 모양이 Convex function인지를 확인하라.
• Convex function: 밥그릇 모양. Gradient descent algorithm을 사용할 때, 초기값으로 무엇을 주더라도 같은 결과가 나온다.

Lecture 4: Multi-variable linear regression

Multi-variable/feature

• represent as matrix
• H(X) = W* X
• H(X) = W transpose * X

Lecture 5-1: Logistic (regression) classification

(Binary) Classification Application

• Spam Detection: Spam(1) or Ham(0)
• Facebook feed: show(1) or hide(0)
• Credit Card Fraudulent Transaction detection: legitimate(0) / fraud(1)

Example Model: Pass(1)/Fail(0) based on study hours

• Linear Regression?
• We know Y is 0 or 1
• Hypothesis can give values large than 1 or less than 0
• => Use Sigmate Fuction!

Sigmoid Function

• Curved in two directions, like the letter "S"

Lecture 5-2: Logistic (regression) classification - cost function & gradient decent

• 어디서 시작하느냐에 따라서 최저점이 달라질 수 있다 -> Local Minimum
• Global Minimum을 찾아야 한다!

New cost function for logistic!

Lecture 6-1: Softax classification - Multinomial classification

Lecture 6-2: Softmax classification - Cross entry cost function & gradient decent

• Cross entropy cost function
• Logistic cost vs cross entropy

Lecture 7-1: Application & Tips - Learning rate, data preprocessing, overfitting

Learning Rate

• Large learning rate: overshooting
• Small learning rate: takes too long, stops at local minimum
• Try several learning rates
  • Observe the cost function
  • Check it goes down in a reasonable rate

Date preprocessing for gradient descent

• Original data -> Zero-centered data -> Normalized data

Overfitting

• 학습 데이터에 너무 잘 맞는(overfitting) 모델. 실제 데이터나 테스트 데이터를 적용하면 잘 안맞는 경우가 발생.
• Solutions for overfitting
  • More training data
  • Reduce the number of features
  • Regularization

Regularization

• Let's not have too big numbers in the weight

Lecture 7-2: Application & Tips - Learning and test data sets

• Training sets -> Validation sets -> Test sets
• Online learning

Accuracy

• How many of your predictions are correct?
• 95~99%