5 papers
Generalization behavior of OPTQ and the role of regularization
Erin George, Rayan Saab
Large neural networks can be compressed by rounding or "quantizing" their weights to numbers that admit representations with fewer bits. One algorithm for quantization, OPTQ, progr…
Observational Multiplicity
Erin George, Deanna Needell, Berk Ustun
Many prediction tasks can admit multiple models that can perform almost equally well. This phenomenon can undermine interpretability and safety when competing models assign conflic…
Towards a Fairer Non-negative Matrix Factorization
Lara Kassab, Erin George, Deanna Needell +3
There has been a recent critical need to study fairness and bias in machine learning (ML) algorithms. Since there is clearly no one-size-fits-all solution to fairness, ML methods s…
Benign overfitting in leaky ReLU networks with moderate input dimension
Kedar Karhadkar, Erin George, Michael Murray +2
The problem of benign overfitting asks whether it is possible for a model to perfectly fit noisy training data and still generalize well. We study benign overfitting in two-layer l…
Training shallow ReLU networks on noisy data using hinge loss: when do we overfit and is it benign?
Erin George, Michael Murray, William Swartworth +1
We study benign overfitting in two-layer ReLU networks trained using gradient descent and hinge loss on noisy data for binary classification. In particular, we consider linearly se…