Often, observational data alone is not enough to uniquely identify a system's causal structure.

Neural Information Processing Systems http://nips.cc/

The inference of causal relationships is a problem of fundamental interest in science. In all fields ofresearch, experiments are systematically performed with the goal ofelucidating the underlying causal dynamics ofsystems.

A common theme in causal inference is learning causal relationships between observed variables, also known as causal discovery. This is usually a daunting task, given the large number of candidate causal graphs and the combinatorial nature of the search space.

Thistask is challenging because the causal graph is unknown and even there may exist latent confounders.

DCBO is useful in scenarios where all causal effects in a graph are changing over time.