|
Date
|
Topic
|
|
Papers
|
Pre-lecture questions
|
Presenters
|
Extra reading
|
|
08/22/2024
|
|
|
Introduction
|
|
|
|
|
08/29/2024
|
Models
|
ResNet
YOLO
|
§ Deep Residual Learning for Image Recognition
§ You Only Look Once: Unified, Real-Time Object Detection
|
|
|
|
|
09/05/2024
|
GCN
Transformer
|
§ Semi-Supervised Classification with Graph Convolutional Networks
§ Attention Is All You Need
|
|
|
|
|
09/12/2024
|
Generative Models
|
VAE
GAN
|
§ Auto-Encoding Variational Bayes
§ Generative Adversarial Nets
|
|
|
|
|
09/19/2024
|
Diffusion Models
|
§ Denoising Diffusion Probabilistic Models
§ Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding
|
|
|
|
|
09/26/2024
|
Model Learning
|
Data Augmentation
|
§ mixup: Beyond Empirical Risk Minimization
§ AutoAugment: Learning Augmentation Strategies from Data
|
|
|
|
|
10/03/2024
|
Transfer Learning
|
§ Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
§ Taskonomy: Disentangling Task Transfer Learning
|
|
|
[Link] [Link]
|
|
10/10/2024
|
Fall Break
|
|
10/17/2024
|
|
Self-supervised Learning
|
§ A Simple Framework for Contrastive Learning of Visual Representations
§ Bootstrap Your Own Latent: A New Approach to Self-Supervised Learning
|
|
|
|
|
10/24/2024
|
Data
|
Privacy & Fairness
Attack & Defense
|
§ A Survey on Bias and Fairness in Machine Learning
§ Explaining and Harnessing Adversarial Examples
|
|
|
|
|
10/31/2024
|
CV
|
CLIP
ViT
|
§ Learning Transferable Visual Models From Natural Language Supervision
§ An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
|
|
|
|
|
11/07/2024
|
LLM
|
BERT
GPT
|
§ BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
§ Language Models are Few-Shot Learners
|
|
|
|
|
11/14/2024
|
Prompting
PEFT
|
§ Making Pre-trained Language Models Better Few-shot Learners
§ LoRA: Low-Rank Adaptation of Large Language Models
|
|
|
|
|
11/21/2024
|
Multimodal
|
Multimodal
|
§ Foundations & Trends in Multimodal Machine Learning: Principles, Challenges, and Open Questions
§ Unified-IO 2: Scaling Autoregressive Multimodal Models with Vision, Language, Audio, and Action
|
|
|
|
|
11/28/2024
|
Thanksgiving
|
|
12/05/2024
|
Course Project Presentation & Demo
|
