Schedule

*This schedule is evolving and will change based on your interests and how the class is progressing. See here for a list of topics and related papers.
Lec. Date Topic Slides Notes
1
Sept. 9
 Introduction Lecture01
2
Sept. 16
 Neural Networks Lecture02 Indicate your preferred papers for in-class presentation due Sept. 16
3
Sept. 23
 Convolutional Neural Networks Lecture03 Assignment 1 is posted on Brightspace. Due on Oct. 8. Tutorial1(PyTorch), Tutorial2(Python).
4
Sept. 30
 Recurrent Neural Networks Lecture04
5
Oct. 7
 Language Models, Transformers
  • Attention Is All You Need (Thomas, Slides)
  • Improving language understanding by generative pre-training (Mozhan, Slides)
  • BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding (Gbemileke, Slides)
  • LoRA: Low-Rank Adaptation of Large Language Models (Muhammad, Slides)
Oct. 14
 Thanksgiving No class Project Proposal due Oct. 12.
Oct. 21
 Fall Break No class Assignment 2 is posted. Due Nov 5
6
Oct. 28
 Vision Transformers, Vision-Language models, Retrieval Augmented Generation
  • An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (Braeden, Slides)
  • Learning transferable visual models from natural language supervision (Ahsan, Slides)
  • Segment Everything Everywhere All at Once (Zeping, Slides)
  • Dense Passage Retrieval for Open-Domain Question Answering (Minjie, Slides)
Lecture06-GANs-Preliminaries
7
Nov. 4
 Generative Adversarial Networks, Diffusion Lecture07-GANs
  • Image-to-image translation with conditional adversarial networks (Sina, Slides)
  • Unpaired image-to-image translation using cycle-consistent adversarial networks (Cristina, Slides)
8
Nov. 11
 Generative Adversarial Networks, Zero-shot Learning
  • A style-based generator architecture for generative adversarial networks (Rowan, Slides)
  • Denoising Diffusion Probabilistic Models (Chintan, Slides)
  • InstructPix2Pix: Learning to Follow Image Editing Instructions(Saeid, Slides)
  • Attribute-based classification for zero-shot visual object categorization (Vishal, Slides)
  • Feature Generating Networks for Zero-Shot Learning (Aziz, Slides)
9
Nov. 18
 Few-shot Learning, Transfer Learning, Domain Adaptation
  • Large Language Models are Zero-Shot Reasoners (Kishore, Slides)
  • Prototypical networks for few-shot learning (Akash, Slides)
  • Learning without Forgetting (Bill (Zhihao), Slides)
  • Transferring Knowledge from Large Foundation Models to Small Downstream Models (Luke, Slides)
  • Cycada: Cycle-consistent Adversarial Domain Adaptation (Ghazaleh, Slides)
10
Nov. 25
 Self-supervised Learning, Deep Clustering
  • A Simple Framework for Contrastive Learning of Visual Representations (Youssef F)
  • Momentum Contrast for Unsupervised Visual Representation Learning (Yousef Y)
  • Masked Autoencoders Are Scalable Vision Learners (Youssef M)
  • Towards K-means-friendly Spaces: Simultaneous Deep Learning and Clustering (Farshid)
  • BERTopic: Neural topic modeling with a class-based TF-IDF procedure (Andrea)
11
Dec. 2
 Interpretable AI
  • Deep Learning for Case-Based Reasoning through Prototypes: A Neural Network that Explains Its Predictions (Connor)
  • This Looks Like That: Deep Learning for Interpretable Image Recognition (Tariq)
  • Grad-cam: Visual explanations from deep networks via gradient-based localization (Adam)
  • Why Should I Trust You?: Explaining the Predictions of Any Classifier (Osama)
  • A unified approach to interpreting model predictions (Omar)
  • Post Hoc Explanations of Language Models Can Improve Language Models (Hoda)
    		•
    12
    Dec. 6
    		 Final Project Presentations
    		Project Reports due on December 10