Generative Path-Law Jump-Diffusion: Sequential MMD-Gradient Flows and Generalisation Bounds in Marcus-Signature RKHS

By Daniel Bloch

Published 2026-04-06

Everscope rating

1379.8

Relevance to quantitative trading

8 / 10

Implementation complexity

10 / 10

Reproducibility

1 / 5

About this paper

Methodology: Anticipatory Neural Jump-Diffusion (ANJD) Flow with Sequential MMD-Gradient Flows in Marcus-Signature RKHS. Problem types: Generative Modeling, Time Series Forecasting, Density Estimation, Optimization, Structured Prediction, Sequence-to-Sequence Learning, Risk Management.

arXiv:2604.05008 · Paper rankings

Open the interactive Everscope explorer for full analysis, charts, and paper battles.