Date
|
Topic
|
Lecturers
|
Materials
|
|
04/04/23
|
Intro to Large Language & Vision Models
- What is AI and why pursue it
- Weak vs Strong AI
- Intro to Large Models for vision and language
- AI4Science applications
- Current LLVMs and failures
- Class overview, homeworks, grading policy
|
|
- Slides - Part A by Pietro: pdf
- Slides - Part B by Georgia: pdf
|
|
04/06/23
|
Brief Recap on MLPs
- Definition of MLPs
- Backpropagation
- Stochastic Gradient Descent
- Momentum (paper)
- Adam (paper)
|
|
|
|
04/11/23
|
Brief Recap on CNNs
|
|
|
|
04/13/23
|
Transformers - Part A
|
|
|
|
04/18/23
|
Transformers - Part B
|
|
|
|
04/20/23
|
Self-Supervised Learning
|
|
|
|
04/25/23
|
Object Recognition at Scale
|
|
|
|
04/27/23
|
Large Vision Models for Segmentation
|
|
|
|
05/02/23
|
Generative Models I
- Intro to Generative Models & Autoregressive Models
- GPT (GPT,
GPT2, GPT3)
- PixelCNN (paper)
- WaveNet (paper)
|
|
|
|
05/04/23
|
Generative Models II
- Variational Autoencoders
- Tutorial on VAEs (paper)
- VQ-VAE (paper)
|
|
- Georgia's slides: pdf
- Assignment 3: 🤖 nanoGPT -- main
& code.
|
|
05/09/23
|
Generative Models III
- Diffusion Models
- DDPMs (paper)
- Denoising Score Matching (paper)
|
|
|
|
05/11/23
|
Generative Models III-1/2
- More on Diffusion Models
- Latent Diffusion (paper)
- Classifier-free Guidance (paper)
|
|
|
|
05/16/23
|
Scalable Vision Foundation Models
|
|
|
|
05/18/23
|
Common Sense: The Dark Matter of Language Intelligence
|
|
- Yejin's slides: pdf
- Jack's slides: pdf
|
|
05/23/23
|
GANs & Unified Vision and Language Models
|
|
- Pietro's slides: pdf
- Georgia's slides: pdf
- Assignment 4: Diffusion 🎨 -- main & code
|
|
05/25/23
|
Stable Diffusion & Image Generation
|
|
|
|
05/30/23
|
Ethics in AI
|
|
|
|
06/01/23
|
Alignment
- Intro to RL
- REINFORCE, PPO
- InstructGPT (paper)
|
|
|
