Advanced Machine Learning

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	Deadlines
1	Jan. 11	Introduction	Lecture01
2	Jan. 18	Neural Networks	Lecture02	~~Indicate your prefered papers for in-class presentation due~~
3	Jan. 25	Convolutional Neural Networks	Lecture03	~~Assignment 1 is out. Due on Feb 7. pdf, Code, Tutorial~~
4	Feb. 1	Recurrent Neural Networks	Lecture04 Attention Is All You Need (Dimo, Slides)
5	Feb. 8	Generative Adversarial Networks (GANs)	Lecture05 Image-to-Image Translation with Conditional Adversarial Networks (Ao, slides) Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks (Ibrahim, slides)	Project Proposal due. Assignment 2 is out. Due on March 8. pdf, Code
	Feb. 15	Winter Break	No class
6	Feb. 22	Generative Adversarial Networks (GANs)	A style-based generator architecture for generative adversarial networks (Kuntal&Ishtiaque, slides) Stargan: Unified generative adversarial networks for multi-domain image-to-image translation (Fardad, slides) Stargan v2: Diverse image synthesis for multiple domains (Mohammad H., slides) Training Generative Adversarial Networks with Limited Data (Daniel&Mohsen, slides) Few-Shot Unsupervised Image-to-Image Translation (Hadi, slides) Semantic Image Synthesis with Spatially-Adaptive Normalization (Aidan, slides)
7	March 1	Few-shot Learning, Transfer Learning	Siamese neural networks for one-shot image recognition (Ishtiaque&Kuntal, slides) FaceNet: A Unified Embedding for Face Recognition and Clustering (Julian, slides) Matching Networks for One Shot Learning (Soroush, slides) Prototypical networks for few-shot learning (Bin, slides) Learning to Compare : Relation Network for Few-Shot Learning (Ummey, slides) Learning without Forgetting (Iman, slides)
8	March 8	Transfer Learning, Domain Adaptation, Zero-shot Learning	Taskonomy: Disentangling Task Transfer Learning (Geetika, slides) Rethinking Pre-training and Self-training (Wei, slides) A Hybrid Instance-based Transfer Learning Method (Rifah, slides) Attribute-Based Classification for Zero-Shot Visual Object Categorization (Afsoon, slides) Semantic Autoencoder for Zero-Shot Learning (Lakshmi, slides) Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders (Keerthana, slides)
9	March 15	Multi-view Learning, Deep Clustering	Sharable and Individual Multi-View Metric Learning (Mohsen&Daniel, slides) Unsupervised deep embedding for clustering analysis (Aditya, slides) SpectralNet: Spectral Clustering using Deep Neural Networks (Yanran, slides) Why Should I Trust You?: Explaining the Predictions of Any Classifier (Zixun, slides)
10	March 22	Interpretable AI	Anchors: High-Precision Model-Agnostic Explanations (Galen&Donglin, slides) Sanity Checks for Saliency Maps (Donglin&Galen, slides) On the (im) possibility of fairness (Mario, slides) Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (Calvin, slides) What Does BERT Look at? An Analysis of BERT’s Attention (Macarious, slides)
11	March 29	Final Project Presentation
12	Apr. 5	Final Project Presentation
				Project Reports due on April 7