Schedule.md

Date	Topic	Readings	Speaker(s)	Weekly Check-in	Notes
8/29	Intro and Logistics	None	Zack Lipton	None	None
9/12	Label and Covariate Shift	Required: Detecting and Correcting for Label Shift with Black Box Predictors Mixture Proportion Estimation and PU Learning: A Modern Approach Optional: Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shift What is the Effect of Importance Weighting in Deep Learning? Domain Adaptation under Open Set Label Shift Unsupervised Learning Under Latent Label Shift Background: On Causal and Anticausal Learning Domain Adaptation under Target and Conditional Shift Improving predictive inference under covariate shift by weighting the log-likelihood function	Zack Lipton	Week 1 ddl: 09/11 11:59PM	notes
9/19	Deep Learning Heuristics	Required: Domain-Adversarial Training of Neural Networks FixMatch Connect, not collapse Also Covered: TENT: Fully Test-Time Adaptation by Entropy Minimization On Learning Invariant Representations for Domain Adaptation Semi-supervised Models are Strong Unsupervised Domain Adaptation Learners mixup: Beyond Empirical Risk Minimization Improve Unsupervised Domain Adaptation with Mixup Training Domain Adaptation with Asymmetrically-Relaxed Distribution Alignment Auxiliary: In Search of Lost Domain Generalization Learning to Generalize: Meta-Learning for Domain Generalization Background: Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks Regularization With Stochastic Transformations and Perturbations for Deep Semi-Supervised Learning Deep Domain Confusion: Maximizing for Domain Invariance	Xueying Ding, Prince Wang, Manley Roberts	Week 2 ddl: 09/19 11:59AM	notes
9/26	Adaptation under Causally Structured Shifts	Required: Domain Adaptation by Using Causal Inference to Predict Invariant Conditional Distributions Domain adaptation under structural causal models Background: Transportability of causal and statistical relations Causal inference and the data-fusion problem Optional General identifiability with arbitrary surrogate experiments Joint Causal Inference from Multiple Contexts Multi-source domain adaptation: A causal view Transporting causal mechanisms for unsupervised domain adaptation Domain adaptation with conditional transferable components Causal generative domain adaptation networks	Beomjo Park, Daniel Jeong	Week 3 ddl: 09/23 1:00PM	notes
10/3	Domain Generalization	Required: Invariant Causal Prediction Invariant Risk Minimization In Search of Lost Domain Generalization Optional: The Risks of Invariant Risk Minimization Learning to Generalize: Meta-Learning for Domain Generalization	Sameer Jain, Pranav Mani	Week 4 ddl: 10/3 12:00pm	notes
10/10	Adversarial Examples	Required: Explaining and Harnessing Adversarial Examples Certified Defenses against Adversarial Examples Background: Intriguing properties of neural networks Optional: Certified Adversarial Robustness via Randomized Smoothing Towards deep learning models resistant to adversarial attacks Robust Physical-World Attacks on Deep Learning Visual Classification Evaluating Robustness of Neural Networks with Mixed Integer Programming	Dhananjay Ashok, Olivier Filion	Week 5 ddl: 10/10 11:59AM	notes
10/17	FALL BREAK/NO CLASS			Week 6	None
10/24	Online Learning/Gradual Shift	Required: Understanding Self-Training for Gradual Domain Adaptation Optimal Dynamic Regret in Exp-Concave Online Learning Background: Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Data Understanding Gradual Domain Adaptation: Improved Analysis, Optimal Path and Beyond Gradual Domain Adaptation without Indexed Intermediate Domains Continuous manifold based adaptation for evolving visual domains Non-Stationary Stochastic Optimization	External speakers: Yu-Xiang Wang, Ananya Kumar	Week 7 ddl:10/24 12pm	notes
10/31	Conformal Inference, Sequential Testing	Required: Adaptive Conformal Inference Under Distribution Shift Conformal Prediction Beyond Exchangeability Background: A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantification Optional: Conformal Prediction Under Covariate Shift Conformal Inference for Online Prediction with Arbitrary Distribution Shifts Adaptive Conformal Predictions for Time Series Testing for Outliers with Conformal p-values Retrain or not retrain: conformal test martingales for change-point detection	Aleksandr Podkopaev, Aditya Gangrade	Week 8 ddl: 10/31 12pm	notes
11/7	Feedback-driven Shifts (Recommender Systems)	Required: Estimating and Penalizing Induced Preference Shifts in Recommender Systems Recommendations as Treatments: Debiasing Learning and Evaluation Optional: Recommendations and User Agency: The Reachability of Collaboratively-Filtered Information Degenerate Feedback Loops in Recommender Systems Feedback Loop and Bias Amplification in Recommender Systems	Thomson Yen, Mel Andrews	Week 9 ddl: 11/7 12pm	notes
11/14	Feedback-driven Shifts (Strategic Classification)	Required: How do classifiers induce agents to invest effort strategically? Strategic Classification is Causal Modeling in Disguise Optional Performative Prediction A Reduction of Imitation Learning and Structured Prediction to No-Regret Online Learning Background Strategic classification	John Long, Michael Agaby	Week 10 ddl: 11/14 12pm	None
11/21	Fairness and Distribution Shifts	Required: Retiring Adult: New Datasets for Fair Machine Learning Delayed Impact of Fair Machine Learning Optional: Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness Background: Mitigating Unwanted Biases with Adversarial Learning 50 Years of Test (Un)fairness: Lessons for Machine Learning	Lingwei Cheng, Mohsen Ferdosi	Week 11 ddl: 11/21 12pm	notes
11/28	No class for NeurIPS			None	None
12/5	Distribution Shifts in the Wild	Required: Source free domain adaptation for medical image segmentation with fourier style mining The Effect of Natural Distribution Shift on Question Answering Models Optional: Unbiased Look at Dataset Bias A Word is Worth A Thousand Dollars: Adversarial Attack on Tweets Fools Stock Prediction CT-GAN: Malicious Tampering of 3D Medical Imagery using Deep Learning	Nicholas Chen, Degan Hao	Week 13	None