DualOpt decouples optimization by using real-time layer-wise weight decay for scratch training and weight rollback for fine-tuning to improve convergence, generalization, and reduce knowledge forgetting.
Split-brain autoencoders: Unsu- pervised learning by cross-channel prediction,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Neural Network Optimization Reimagined: Decoupled Techniques for Scratch and Fine-Tuning
DualOpt decouples optimization by using real-time layer-wise weight decay for scratch training and weight rollback for fine-tuning to improve convergence, generalization, and reduce knowledge forgetting.