The preconditioning exponent p in Adam-like optimizers systematically shifts the relative update ratio between weight and bias parameters, changing which samples the model fits first.
Advances in neural information processing systems30(2017)
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
Introduces ML-FOP-SOAP optimizer using Fisher-Orthogonal Projection and hierarchical folding to mitigate modality competition in multimodal autoregressive training, reporting gains over AdamW on Janus and Emu3.
citing papers explorer
-
Information Allocation Dynamics in Neural Network Optimization
The preconditioning exponent p in Adam-like optimizers systematically shifts the relative update ratio between weight and bias parameters, changing which samples the model fits first.
-
Second-Order Multi-Level Variance Correction for Modality Competition in Multimodal Models
Introduces ML-FOP-SOAP optimizer using Fisher-Orthogonal Projection and hierarchical folding to mitigate modality competition in multimodal autoregressive training, reporting gains over AdamW on Janus and Emu3.