Abstract:We consider backward stochastic differential equations (BSDEs) with mean-field and McKean-Vlasov interactions in their generators in a general setting, where the drivers are square-integrable martingales, with a focus on the independent increments case, and the filtrations are (possibly) stochastically discontinuous. In other words, we consider discrete- and continuous-time systems of mean-field BSDEs and McKean-Vlasov BSDEs in a unified setting. We provide existence and uniqueness results for these BSDEs using new a priori estimates that utilize the stochastic exponential. Then, we provide propagation of chaos results for systems of particles that satisfy BSDEs, i.e. we show that the asymptotic behaviour of the solutions of mean-field systems of BSDEs, as the multitude of the systems grows to infinity, converges to I.I.D solutions of McKean-Vlasov BSDEs. We introduce a new technique for showing the backward propagation of chaos, that makes repeated use of the a priori estimates, inequalities for the Wasserstein distance and the ``conservation of solutions'' under different filtrations, and does not demand the solutions of the mean field systems to be exchangeable or symmetric. Finally, we deduce convergence rates for the propagation of chaos, under advanced integrability conditions on the solutions of the BSDEs.

What problem does this paper attempt to address?

### Problems the paper attempts to solve This paper mainly discusses the existence, uniqueness and chaos propagation problems of two types of backward stochastic differential equations (BSDEs), specifically involving generators with mean - field and McKean - Vlasov interactions. The authors study these equations under a general setting and pay special attention to the unified treatment of discrete - time and continuous - time systems in the case of independent increments. #### 1. **Existence and Uniqueness Problems** The paper first addresses the existence and uniqueness problems of two types of BSDEs: - **McKean - Vlasov BSDEs**: The generator of this type of equation depends on the distribution of the solution. - **Mean - field system BSDEs**: The generator of this type of equation depends on the empirical measure of an N - particle system. By introducing a new a priori estimation method, especially by using the stochastic exponential, the authors prove the existence and uniqueness of these two types of BSDEs under broader conditions. #### 2. **Chaos Propagation Problem** The core contribution of the paper lies in the study of the backward propagation of chaos in the mean - field BSDEs system. Specifically, the authors show that when the number of particles \( N \) in the system tends to infinity, the solution of the mean - field BSDEs system converges to the solution of the independently and identically distributed McKean - Vlasov BSDEs. This indicates that in large systems, the interactions between individuals can be simplified to a deterministic distribution. To prove this result, the authors introduce a new technique, which repeatedly uses a priori estimates, Wasserstein distance inequalities and the "conservativeness" of solutions under different filtrations, without the need to assume the exchangeability or symmetry of solutions. #### 3. **Convergence Rate** In addition, the paper also provides an estimate of the convergence rate of chaos propagation. Under certain advanced integrability conditions, the authors derive a specific convergence rate. This further verifies the practical application value of the theoretical results. ### Summary The main objectives of this paper are: - **To prove the existence and uniqueness of two types of BSDEs**: including McKean - Vlasov BSDEs and mean - field system BSDEs. - **To study the backward chaos propagation phenomenon**: to show the behavior of the solution of the mean - field BSDEs system in the large - system limit. - **To provide an estimate of the convergence rate**: to give a specific convergence rate of chaos propagation under specific conditions. Through these studies, the authors provide a powerful tool for understanding and analyzing the asymptotic behavior of complex systems, especially having important applications in fields such as financial mathematics and statistical physics.

Existence, uniqueness and propagation of chaos for general McKean-Vlasov and mean-field BSDEs

Propagation of chaos for moderately interacting particle systems related to singular kinetic Mckean-Vlasov SDEs

Strong Convergence of Propagation of Chaos for McKean–Vlasov SDEs with Singular Interactions

Quantitative weak propagation of chaos for stable-driven McKean-Vlasov SDEs

Propagation of chaos for mean field Schrödinger problems

Set-valued propagation of chaos for controlled path-dependent McKean-Vlasov SPDEs

Existence, uniqueness and ergodicity for McKean-Vlasov SDEs under distribution-dependent Lyapunov conditions

Wellposedness, exponential ergodicity and numerical approximation of fully super-linear McKean--Vlasov SDEs and associated particle systems

Sticky nonlinear SDEs and convergence of McKean-Vlasov equations without confinement

Mean-field forward and backward SDEs with jumps. Associated nonlocal quasi-linear integral-PDEs

The random periodic solutions for McKean-Vlasov stochastic differential equations

A note on Lévy-driven McKean-Vlasov SDEs under monotonicity

Quantitative Propagation of Chaos for Singular Interacting Particle Systems Driven by Fractional Brownian Motion

Uniform-in-time estimates for mean-field type SDEs and applications

Entropic propagation of chaos for mean field diffusion with $L^p$ interactions via hierarchy, linear growth and fractional noise

Well-posedness and propagation of chaos for jump-type McKean-Vlasov SDEs with irregular coefficients

A BSDEs approach to pathwise uniqueness for stochastic evolution equations

Stability of McKean–Vlasov stochastic differential equations and applications

Global Dynamics of McKean-Vlasov SDEs via Stochastic Order

Non-asymptotic convergence rates for mean-field games: weak formulation and McKean-Vlasov BSDEs

Entropy on the Path Space and Application to Singular Diffusions and Mean-field Models