Generative model for rough volatility: log-signatures + a learned Besov-wavelet decoder reconstruct high-frequency texture via differentiable IDWT. Pluggable MLP/attention/transformer backbones, scale-weighted wavelet loss, and a 5-dataset multi-domain registry (fBM, rough Bergomi, Burgers turbulence, CHB-MIT EEG, ESC-50 audio).
audio benchmark deep-learning pytorch eeg generative-model quantitative-finance turbulence wavelets rough-volatility fractional-brownian-motion diffusion-models time-series-generation rough-bergomi signature-method besov-space
-
Updated
May 3, 2026 - Jupyter Notebook