Full Stack Deep Learning - Spring 2021 강의로 스터디 진행
- 모든 방식은 멤버들과의 합의에 따라 바뀔 수 있음
- 매주 일요일 Gooogle Meet 로 스터디 진행
- 매 회주 스터디가 끝난 후 회고를 진행하고, 진행했던 내용을 Log 로 남김 (멤버공개)
| day | Topic | Role | Tag |
|---|---|---|---|
| 21.05.21 | 1.Fundamental | * Presenter : 엄다연 * Lecture1 : 조준우 * Lab1 : 허원재 * git-manager: 여상엽 |
# Neural Networks # Universality # Learning Problems # Loss Function # Backpropagation # CUDA # Setup # Train MLP model MNIST # Tensorflow # Pytorch |
| 21.05.30 | 2.CNNs | * Presenter : 허원재 * Lecture2A : 조준우 * Lecture2B : 여상엽 * Lab : 엄다연 |
# Conv # Stride # Padding # Receptive Field # 1x1 Conv # AlexNet # ZFNet # VGGNet # GoogLeNet # RestNet # Localization # Detection # Overfit # Adversarial Attacks # Train CNN Model EMNIST # synthetic EMNIST lines |
| 21.06.20 | 3.RNNs | * Presenter : 조준우 * Lecture3 : 여상엽 * Lab3 : 엄다연 * git-manager: 허원재 |
# sequence problem # disadvantage of Feedforward networks # gradient vanishing # LSTMs # forget gate #input gate # problems of LSTMs # resNet #attention #bidirectionality #CTC loss #convNet |
| 21.06.27 | 4.Transformers | * Presenter : 여상엽 * Lecture4 : 엄다연 * Lab4 : 허원재 * git-manager: 조준우 |
#Transfer Learning in Computer Vision #fine-tuning #Embeddings and Language Models #word #vector #one-hot encoding #pre-training #ELMO #ULMFit #NLP #SQuAD #SNLI #GLUE #Transformers #Attention #BERT #GPT/GPT-2/GPT-3 # DistillBERT #T5 #etc |
| 21.07.04 | 5.ML Projects | * Presenter : 엄다연 * Lecture5 : 허원재 * 조준우 * 여상엽 |
#~~~~ |