Deep Learning and Elicitability for McKean-Vlasov FBSDEs With Common Noise

Name: Deep Learning and Elicitability for McKean-Vlasov FBSDEs With Common Noise
Uploaded: 2026-10-15T20:07:02.427688+00:00
Duration: 487 s
Description: This work introduces a novel numerical method combining Picard iterations, elicitability, and deep learning to solve complex McKean-Vlasov forward-backward stochastic differential equations with common noise. The approach uses neural networks parameterized by elicitable scores to approximate difficult forward and backward processes without relying on expensive nested Monte Carlo simulations.

This work introduces a novel numerical method combining Picard iterations, elicitability, and deep learning to solve complex McKean-Vlasov forward-backward stochastic differential equations with common noise. The approach uses neural networks parameterized by elicitable scores to approximate difficult forward and backward processes without relying on expensive nested Monte Carlo simulations.

Open the full episode