SGD is reformulated via a master equation from discrete updates, producing a discrete Fokker-Planck equation that predicts non-stationary variance growth proportional to learning rate in flat Hessian directions.
Deep learning
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
DPF-GFD uses complementary frequency filtering on the original graph and a similarity graph to produce more discriminative node embeddings for fraud detection under high heterophily and class imbalance.
citing papers explorer
-
Why SGD is not Brownian Motion: A New Perspective on Stochastic Dynamics
SGD is reformulated via a master equation from discrete updates, producing a discrete Fokker-Planck equation that predicts non-stationary variance growth proportional to learning rate in flat Hessian directions.
-
Graph-Based Fraud Detection with Dual-Path Graph Filtering
DPF-GFD uses complementary frequency filtering on the original graph and a similarity graph to produce more discriminative node embeddings for fraud detection under high heterophily and class imbalance.