Eva L. Dyer

LatentDR: Improving Model Generalization Through Sample-Aware Latent Degradation and Restoration

2023 · arXiv (Cornell University)

Despite significant advances in deep learning, models often struggle to generalize well to new, unseen domains, especially when training data is limited. To address this challenge, we propose a novel approach for distribution-aware latent augmentation …

Balanced Data, Imbalanced Spectra: Unveiling Class Disparities with Spectral Imbalance

2024 · arXiv (Cornell University)

Classification models are expected to perform equally well for different classes, yet in practice, there are often large gaps in their performance. This issue of class bias is widely studied in cases of datasets with …