Khoat Than

LCA: Local Classifier Alignment for Continual Learning

A Tight Error Bound for Deep Learning via Distribution and Loss Complexity

A Non-vacuous Test Error Guarantee for Deep Learning without Altering the Model

SUSI: Semi-Structured Pruning for LLMs via Differentiable Subset Sampling

DPaI: Differentiable Pruning at Initialization with Node-Path Balance Principle

Non-vacuous Bounds for the test error of Deep Learning without any change to the trained models

Boosting Multiple Views for pretrained-based Continual Learning

Provably Improving Generalization of Few-shot models with Synthetic Data