딥러닝 / CS231N / LECTURE
LECTURE
-
CS231n Lecture 18 - Human-Centered AI
human-centered computer vision, fairness, privacy, healthcare, assistive robotics
-
CS231n Lecture 17 - Robot Learning
reinforcement learning, model-based planning, imitation learning, vision-language-action models
-
CS231n Lecture 16 - Vision and Language
CLIP, multimodal language models, grounding, model chaining
-
CS231n Lecture 15 - 3D Vision
3D representations, neural fields, NeRF, Gaussian splatting
-
CS231n Lecture 14 - Generative Models 2
Generative adversarial networks, rectified flow, latent diffusion
-
CS231n Lecture 13 - Generative Models 1
Maximum likelihood, autoregressive models, variational autoencoders
-
CS231n Lecture 12 - Self-Supervised Learning
Pretext task, MAE, contrastive representation learning
-
CS231n Lecture 11 - Large-Scale Distributed Training
GPU cluster의 memory, communication, parallelism
-
CS231n Lecture 10 - Video Understanding
Spatiotemporal representation과 action recognition
-
CS231n Lecture 9 - Object Detection, Image Segmentation, and Visualization
Dense prediction과 모델 시각화
-
CS231n Lecture 8 - Attention and Transformers
Attention 연산과 Transformer 구조
-
CS231n Lecture 7 - Recurrent Neural Networks
RNN 구조와 sequence modeling
-
CS231n Lecture 6 - Training CNNs and CNN Architectures
CNN 구성 요소와 아키텍처, 초기화, 데이터 증강, 전이 학습, 학습 방법
-
CS231n Lecture 5 - Image Classification with CNNs
CNN 이미지 분류
-
CS231n Lecture 4 - Neural Networks and Backpropagation
Neural Networks, Backprop
-
CS231n Lecture 3 - Regularization and Optimization
Regularization의 목적, L1/L2 정규화, gradient descent, SGD, momentum, RMSProp, Adam과 learning rate schedule
-
CS231n Lecture 2 - Image Classification with Linear Classifiers
이미지 분류 문제, 데이터 기반 접근, Nearest Neighbor와 K-NN, 선형 분류기의 입력 shape와 score 계산 정리
-
CS231n Lecture 1 - Introduction
컴퓨터 비전의 문제 정의, 데이터 기반 접근, 이미지 분류와 딥러닝으로 이어지는 흐름 정리