summary: The authors study the problem of identifying a causal DAG from observational and active (ide closed-loop) interventional data. They lower bound the minimal number of interventions required to orient any graph in terms of the size of the largest cliques in the essential graph of the causal model. pro: - minimal-cost, active causal identification is an important topic - new lower bound on complexity of active identification of a causal model - readily applicable identification algorithm that comes close to achieving the lower bound on a restricted set of causal models cons: - somewhat restricted setting of no unobserved confounders, perfect atomic interventions only - empirical results are somewhat weak meta review: Solid paper on an important topic with novel theoretical results and a practical algorithmic approach.