To understand the nature of the complex behavior of the brain, one important step is to identify "cores" in the brain network, where neurons or brain areas strongly interact with each other. Cores can be considered as essential sub-networks for brain functions. In the last few decades, an information-theoretic approach to identifying cores has been developed. In this approach, many-to-many nonlinear interactions between parts are measured by an information loss function, which quantifies how

more »

... h information would be lost if interactions between parts were removed. Then, a core called a "complex" is defined as a subsystem wherein the amount of information loss is locally maximal. Although identifying complexes can be a novel and useful approach to revealing essential properties of the brain network, its practical application is hindered by the fact that computation time grows exponentially with system size. Here we propose a fast and exact algorithm for finding complexes, called Hierarchical Partitioning for Complex search (HPC). HPC finds complexes by hierarchically partitioning systems to narrow down candidates for complexes. The computation time of HPC is polynomial, which is dramatically smaller than exponential. We prove that HPC is exact when an information loss function satisfies a mathematical property, monotonicity. We show that mutual information is one such information loss function. We also show that a broad class of submodular functions can be considered as such information loss functions, indicating the expandability of our framework to the class. In simulations, we show that HPC can find complexes in large systems (up to several hundred) in a practical amount of time when mutual information is used as an information loss function. Finally, we demonstrate the use of HPC in electrocorticogram recordings from monkeys. HPC revealed temporally stable and characteristic complexes, indicating that it can be reliably utilized to characterize brain networks.