conftrace_

stagewise training

3 papers

Explore in graph

Co-occurring keywords · papers mentioning both

subnetwork training (1) curriculum loss (1) polyak-łojasiewicz condition (1) layer dropping (1) pre-training method (1) label corruption (1) sample selection (1) convergence rate (1) 0-1 loss (1) robust learning (1)

Papers

Efficient stagewise pretraining via progressive subnetworks ICLR 2025 Curriculum Loss: Robust Learning and Generalization against Label Corruption ICLR 2020 Stagewise Training Accelerates Convergence of Testing Error Over SGD NeurIPS 2019