Here you can find a collection of relevant readings for projects going on in the lab. These will be useful starting points to give you some background for joining your new project!

Many of these subjects merge, so you can expect to find papers which bring some of these ideas together, e.g. fairness in privacy-preserving techniques.

There is also the paper-exchange channel in Slack, which people will update with new and interesting papers.

Fairness

Fairness definitions explained: https://fairware.cs.umass.edu/papers/Verma.pdf

The impossibility of fairness: https://arxiv.org/abs/1609.07236

Balanced datasets are not enough: https://arxiv.org/abs/1811.08489

Reducing gender bias amplification using corpus level constraints: https://arxiv.org/abs/1707.09457

Privacy

Federated learning: distributed learning of deep neural networks over multiple agents: https://www.sciencedirect.com/science/article/abs/pii/S1084804518301590

Comprehensive privacy analysis of deep learning: stand-alone and federated learning under passie and active white-box inference attacks: https://arxiv.org/abs/1812.00910

Membership inference attacks against machine learning models: https://arxiv.org/abs/1610.05820

The secret sharer: evaluating and testing unintended memorization in neural networks: https://arxiv.org/abs/1802.08232

Differentially private empirical risk minimization under the fairness lens: https://proceedings.neurips.cc/paper/2021/file/e7e8f8e5982b3298c8addedf6811d500-Paper.pdf

Overlearning reveals sensitive attributes: https://arxiv.org/abs/1905.11742

Interpretability and Explainability

Understanding black-box predictions via influence functions: https://arxiv.org/abs/1703.04730