[Course 2] Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization.W1 - Practical Aspects of Deep Learning

[Coursera] Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization 1주차 강의 (Practical Aspects of Deep Learning)를 수강하며 작성한 필기 노트입니다.

---

1. Setting up your Machine Learning Application

1-1. Train/ Dev / Test sets

 

1-2. Bias / Variance

 

1-3. Basic Recipe for Machine Learning

 

2. Regularizing your Neural Network

반응형

2-1. Regularization

 

2-2. Why Regularization Reduces Overfitting?

 

2-3. Dropout Regularization

 

2-4. Understanding Dropout

 

2-5. Other Regularization Methods

 

3. Setting Up your Optimization Problem

 

3-1. Normalizing Inputs

 

3-2. Vanishing / Exploding Gradients

 

3-3. Weight Initialization for Deep Networks

 

3-4. Numerical Approximation of Gradients

 

3-5. Gradient Checking

 

3-6. Gradient Checking Implementation Notes

 

 

 

---