06. Neural Networks Part 2: Setting up the Data and the Loss
데이터 전처리, 가중치 초기화, batch normalization, 정규화와 손실 함수 설정.
이용학
05. Neural Networks Part 1: Setting up the Architecture
생물학적 뉴런에서 출발한 신경망 모델, 활성화 함수의 종류, 층 구조와 표현력.
04. Backpropagation, Intuitions
연쇄 법칙에 기반한 역전파의 직관적 이해, 계산 그래프와 게이트, 벡터화된 기울기 계산.
03. Optimization: Stochastic Gradient Descent
손실 함수 최적화, 수치적·해석적 기울기 계산, 경사 하강법과 mini-batch SGD.
02. Linear Classification: Support Vector Machine, Softmax
점수 함수와 손실 함수로 이루어진 선형 분류, Multiclass SVM과 Softmax 분류기, 그리고 정규화.
01. Image Classification: Data-driven Approach, k-Nearest Neighbor, train/val/test splits
이미지 분류 문제, 데이터 기반 접근법, k-최근접 이웃 분류기, 하이퍼파라미터 튜닝을 위한 train/val/test 분할.
GIU Essential (Adjectives and Adverbs 085-092)
Unit 085. old/nice/interesting etc. (adjectives) A adjective + noun It’s a nice day today. Laura has got brown eyes. There’s a very old bridge in this village. Do you like Italian fo...
GIU Essential (Determiners and Pronouns 074-084)
Unit 074. this/that/these/those A this (singular) Do you like this picture?? (this picture here) these (plural) These flowers are for you. (thes...
GIU Essential (A and the 065-073)
Unit 065. a/an … A a … = one thing or person: He’s got a camera. She’s waiting for a taxi. It’s a beautiful day. Rachel works in a bank. Can I ask a question? I haven’t got a com...
GIU Essential (Pronouns and possessives 059-064)
Unit 059. I/me he/him they/them etc. A I know Tom. / Tom knows me. We know Tom. / Tom knows us. You know Tom. / Tom knows you. He knows Tom. / Tom knows him. She knows Tom. / Tom k...