A bias-correction framework for stochastic preconditioned optimizers (AdamW, Sophia, Shampoo) using cross-fitted microbatches and delta-method inversion correction yields 0.07-0.15 nat loss reductions on Qwen2.5-0.5B pretraining.
International Conference on Learning Representations , year =
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5representative citing papers
PolarAdamW disentangles spectral control from gauge-equivariance in matrix optimizers, with experiments demonstrating their distinct roles on standard versus symmetry-aware neural networks.
Covariance-aware goodness and auxiliary modules let Forward-Forward training scale to 16-layer networks, achieving 73.01% on ImageNet-100 and 50.30% on Tiny-ImageNet with roughly half the peak memory of backpropagation.
Setting β in balanced Adam to achieve a refresh count R_β ≈1000 based on effective learning horizon T_ES improves validation robustness over fixed-β baselines across 11 vision and language experiments.
citing papers explorer
-
Correcting Stochastic Update Bias in Preconditioned Language Model Optimizers
A bias-correction framework for stochastic preconditioned optimizers (AdamW, Sophia, Shampoo) using cross-fitted microbatches and delta-method inversion correction yields 0.07-0.15 nat loss reductions on Qwen2.5-0.5B pretraining.
-
PolarAdamW: Disentangling Spectral Control and Schur Gauge-Equivariance in Matrix Optimisation
PolarAdamW disentangles spectral control from gauge-equivariance in matrix optimizers, with experiments demonstrating their distinct roles on standard versus symmetry-aware neural networks.
-
Covariance-Aware Goodness for Scalable Forward-Forward Learning
Covariance-aware goodness and auxiliary modules let Forward-Forward training scale to 16-layer networks, achieving 73.01% on ImageNet-100 and 50.30% on Tiny-ImageNet with roughly half the peak memory of backpropagation.
-
Refresh-Scaling the Memory of Balanced Adam
Setting β in balanced Adam to achieve a refresh count R_β ≈1000 based on effective learning horizon T_ES improves validation robustness over fixed-β baselines across 11 vision and language experiments.
- Normalization Equivariance for Arbitrary Backbones, with Application to Image Denoising