| Lec. |
Date |
Topic |
Slides |
Notes |
| 1 |
Sept. 12 |
Introduction |
Lecture01 |
|
| 2 |
Sept. 19 |
Neural Networks |
Lecture02 |
Indicate your preferred papers for in-class presentation due Sept. 18 |
| 3 |
Sept. 26 |
Convolutional Neural Networks |
Lecture03 |
Assignment 1 is posted on Brightspace. Due on Oct 11. Tutorial1(PyTorch), Tutorial2(Python) |
| 4 |
Oct. 3 |
Recurrent Neural Networks |
Lecture04 |
|
|
Oct. 10 |
|
No Class |
Project Proposal due Oct. 12. Assignment 2 is posted. Due Oct 20. |
| 5 |
Oct. 17 |
Language Models, Transformers |
Lecture05
- Improving Language Understanding by Generative Pre-Training (Kaishuo, slides)
- BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding (Sri, slides)
- Prefix-tuning: Optimizing continuous prompts for generation (Shan, slides)
- LoRA: Low-Rank Adaptation of Large Language Models (Mohammad, slides)
|
|
| 6 |
Oct. 18 |
Vision Transformers, Generative Adversarial Networks (GAN) |
- An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (Elmira, slides)
- Learning transferable visual models from natural language supervision (Alex, slides)
Lecture06
|
|
|
Oct. 24 |
Fall Break |
No Class |
|
| 7 |
Oct 31 |
Generative Adversarial Networks, Diffusion |
- Image-to-image translation with conditional adversarial networks (Anthony, slides)
- Unpaired image-to-image translation using cycle-consistent adversarial networks (Feier, slides)
- StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation (Adnan, slides)
- Denoising Diffusion Probabilistic Models (Tianshu, slides)
- BBDM: Image-to-Image Translation with Brownian Bridge Diffusion Models (Alireza, slides)
|
Assignment 3 is posted. Due Nov. 17 |
| 8 |
Nov. 7 |
Few-shot Learning, Zero-shot Learning |
- Semantic autoencoder for zero-shot learning (Kamran, slides)
- Prototypical networks for few-shot learning (Jean-Luc, slides)
- Learning to Compare: Relation Network for Few-Shot Learning (Dawei, slides)
- Large Language Models are Zero-Shot Reasoners (Kaya, slides)
- Few-shot unsupervised image-to-image translation (Shervin, slides)
|
|
| 9 |
Nov. 14 |
Self-supervised Learning, Deep Clustering, Interpretable AI |
- A simple framework for contrastive learning of visual representations (Aidan, slides)
- Bootstrap Your Own Latent A New Approach to Self-Supervised Learning (Benjamin, slides)
- Towards K-means-friendly Spaces: Simultaneous Deep Learning and Clustering (Patrick, slides)
- Why Should I Trust You?: Explaining the Predictions of Any Classifier (Yaqing, slides)
- A Unified Approach to Interpreting Model Predictions (Ryan, slides)
|
|
| 10 |
Nov. 21 |
Interpretable AI |
- Attention is not explanation (Amin, slides)
- Interpretability beyond feature attribution: Quantitative testing with concept activation vectors (Janaki, slides)
- Cause and Effect: Concept-based Explanation of Neural Networks (Rohit, slides)
- This Looks Like That: Deep Learning for Interpretable Image Recognition (Hongbo, slides)
|
|
| 11 |
Nov. 28 |
Final Project Presentations |
|
|
| 12 |
Dec. 5 |
Final Project Presentations |
|
|
|
|
|
|
Project Reports due on December 10 |