Neural network framework for volatility surface approximation and calibration. Supports rough Heston/Bergomi, random grids, multi-regime architectures.

Numba-accelerated Rough Bergomi volatility model for derivatives pricing. Tested on Tesla (TSLA).

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).