Machine Learning

추천 시스템: 조회, 채점, 순위 재지정 (Retrieval, Scoring, Re-ranking)

4 분 소요

Retrieval, Scoring, Re-ranking

추천 시스템: 심층 신경망 (Deep Neural Network)

5 분 소요

Softmax Model, Softmax Training

추천 시스템: 협업 필터링 (Collaborative filtering)

6 분 소요

Basics, Matrix Factorization, Advantages & Disadvantages

추천 시스템: 컨텐츠 기반 필터링 (Content-based filtering)

1 분 소요

Basics, Advantages & Disadvantages

추천 시스템: 후보 생성 (Candidate Generation)

2 분 소요

Overview

추천 시스템이란? (What is Recommendation System?)

2 분 소요

What and Why?, Terminology, Overview

공정성 (Fairness)

6 분 소요

Types of Bias, Identifying Bias, Evaluating for Bias

데이터 종속성 (Data Dependency)

2 분 소요

Raliability, Versioning, Necessity, Correlations, Feedback Loops

정적 vs. 동적 (Static vs. Dynamic)

1 분 소요

Static vs. Dynamic Training and Inference

임베딩 (Embedding)

7 분 소요

Collaborative Filtering, Input Data, Translating, Obtaining

다중 클래스 신경망 (Multi-Class Neural Networks)

2 분 소요

One vs. All, Softmax

신경망 학습 (Training Neural Nets)

1 분 소요

Best Practices

신경망 (Neural Networks)

2 분 소요

Structure : Hidden Layers, Activation Functions

정규화 : 희소성 (Regularization for Sparsity)

2 분 소요

L1 Regularization

분류 (Classification)

7 분 소요

Threshold, True-False, Positive-Negative, Accuracy, Precision, Recall, ROC, AUC, Prediction Bias

로지스틱 회귀 (Logistic Regression)

2 분 소요

Calculating a Probability, Loss and Regularization

정규화 : 단순성 (Regularization for Simplicity)

2 분 소요

L2 Regularization, Lambda

특성 교차 (Feature Crosses)

3 분 소요

Encoding Nonlinearity, Crossing One-Hot Vectors

데이터 표현 (Representation)

9 분 소요

Feature Engineering, Qualities of Good Features, Cleaning Data

데이터 세트 구축 (Splitting Data)

2 분 소요

Training, Validation and Test Dataset

일반화 (Generalization)

2 분 소요

Peril of Overfitting

Loss 감소 시키기(Reducing Loss)

4 분 소요

Iterative Approach, Gradient Desent, Learning Rate

머신러닝 시작하기

2 분 소요

Linear Regression, Training, Loss