Predicting node labels on a given graph is a widely studied problem with...
We consider finding flat, local minimizers by adding average weight
pert...
Multitask learning is widely used in practice to train a low-resource ta...
Suppose there is a spreading process such as an infectious disease
propa...
Graph neural networks are widely used tools for graph prediction tasks.
...
We consider transfer learning approaches that fine-tune a pretrained dee...
High-capacity deep neural networks (DNNs) trained with Empirical Risk
Mi...
Spurious correlations pose a major challenge for robust machine learning...
A widely used algorithm for transfer learning is fine-tuning, where a
pr...
Hard parameter sharing for multi-task learning is widely used in empiric...
We consider the dynamic of gradient descent for learning a two-layer neu...
We investigate multi-task learning approaches that use a shared feature
...
Data augmentation is a powerful technique to improve performance in
appl...