Identifying Nonlinear 1-Step Causal Influences in Presence of Latent Variables

We propose an approach for learning the causal structure in stochastic\ndynamical systems with a $1$-step functional dependency in the presence of\nlatent variables. We propose an information-theoretic approach that allows us\nto recover the causal relations among the observed variables as long as the\nlatent variables evolve without exogenous noise. We further propose an\nefficient learning method based on linear regression for the special sub-case\nwhen the dynamics are restricted to be linear. We validate the performance of\nour approach via numerical simulations.\n