Towards Identifiability of Interventional Stochastic Differential Equations
This addresses a foundational challenge in causal inference for continuous-time dynamical systems, with potential applications in fields like gene regulatory dynamics.
The paper tackles the problem of uniquely recovering parameters in stochastic differential equations (SDEs) from observational data under interventions, providing the first provable bounds for identifiability with tight results for linear SDEs and upper bounds for nonlinear cases.
We study identifiability of stochastic differential equations (SDE) under multiple interventions. Our results give the first provable bounds for unique recovery of SDE parameters given samples from their stationary distributions. We give tight bounds on the number of necessary interventions for linear SDEs, and upper bounds for nonlinear SDEs in the small noise regime. We experimentally validate the recovery of true parameters in synthetic data, and motivated by our theoretical results, demonstrate the advantage of parameterizations with learnable activation functions in application to gene regulatory dynamics.