An official website of the United States government
Abstract In a complex network, the core component with interesting structures is usually hidden within noninformative connections. The noises and bias introduced by the noninformative component can obscure the salient structure and limit many network modeling procedures’ effectiveness. This paper introduces a novel core–periphery model for the noninformative periphery structure of networks without imposing a specific form of the core. We propose spectral algorithms for core identification for general downstream network analysis tasks under the model. The algorithms enjoy strong performance guarantees and are scalable for large networks. We evaluate the methods by extensive simulation studies demonstrating advantages over multiple traditional core–periphery methods. The methods are also used to extract the core structure from a citation network, which results in a more interpretable hierarchical community detection.
more »
« less
Hierarchical core-periphery structure in networks
We study core-periphery structure in networks using inference methods based on a flexible network model that allows for traditional onion-like cores within cores, but also for hierarchical tree-like structures and more general non-nested types of structure. We propose an efficient Monte Carlo scheme for fitting the model to observed networks and report results for a selection of real-world data sets. Among other things, we observe an empirical distinction between networks showing traditional core-periphery structure with a dense core weakly connected to a sparse periphery, and an alternative structure in which the core is strongly connected both within itself and to the periphery. Networks vary in whether they are better represented by one type of structure or the other. We also observe structures that are a hybrid between core-periphery structure and community structure, in which networks have a set of non-overlapping cores that correspond roughly to communities, surrounded by a single undifferentiated periphery. Computer code implementing our methods is available.
more »
« less
- Award ID(s):
- 2005899
- PAR ID:
- 10537879
- Publisher / Repository:
- American Physical Society
- Date Published:
- Journal Name:
- Physical Review E
- Volume:
- 108
- Issue:
- 2
- ISSN:
- 2470-0045
- Page Range / eLocation ID:
- 024311
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Pruning neural networks at initialization (PaI) has received an upsurge of interest due to its end-to-end saving potential. PaI is able to find sparse subnetworks at initialization that can achieve comparable performance to the full networks. These methods can surpass the trivial baseline of random pruning but suffer from a significant performance gap compared to post-training pruning. Previous approaches firmly rely on weights, gradients, and sanity checks as primary signals when conducting PaI analysis. To better understand the underlying mechanism of PaI, we propose to interpret it through the lens of the Ramanujan Graph - a class of expander graphs that are sparse while being highly connected. It is often believed there should be a strong correlation between the Ramanujan graph and PaI since both are about finding sparse and well-connected neural networks. However, the finer-grained link relating highly sparse and connected networks to their relative performance (i.e., ranking of difference sparse structures at the same specific global sparsity) is still missing. We observe that not only the Ramanujan property for sparse networks shows no significant relationship to PaI’s relative performance, but maximizing it can also lead to the formation of pseudo-random graphs with no structural meanings. We reveal the underlying cause to be Ramanujan Graph’s strong assumption on the upper bound of the largest nontrivial eigenvalue (µˆ) of layers belonging to highly sparse networks. We hence propose Iterative Mean Difference of Bound (IMDB) as a mean to relax the µˆ upper bound. Likewise, we also show there exists a lower bound for µˆ, which we call the Normalized Random Coefficient (NaRC), that gives us an accurate assessment for when sparse but highly connected structure degenerates into naive randomness. Finally, we systematically analyze the behavior of various PaI methods and demonstrate the utility of our proposed metrics in characterizing PaI performance. We show that subnetworks preserving better the IMDB property correlate higher in performance, while NaRC provides us with a possible mean to locate the region where highly connected, highly sparse, and non-trivial Ramanujan expanders exist. Our code is available at: https://github.com/VITA-Group/ramanujan-on-pai.more » « less
-
Colloidal nanocrystals consist of an inorganic crystalline core with organic ligands bound to the surface and naturally self-assemble into periodic arrays known as superlattices. This periodic structure makes superlattices promising for phononic crystal applications. To explore this potential, we use plane wave expansion methods to model the phonon band structure. We find that the nanoscale periodicity of these superlattices yield phononic band gaps with very high center frequencies on the order of 10 2 GHz. We also find that the large acoustic contrast between the hard nanocrystal cores and the soft ligand matrix lead to very large phononic band gap widths on the order of 10 1 GHz. We systematically vary nanocrystal core diameter, d , nanocrystal core elastic modulus, E NC core , interparticle distance ( i.e. ligand length), L , and ligand elastic modulus, E ligand , and report on the corresponding effects on the phonon band structure. Our modeling shows that the band gap center frequency increases as d and L are decreased, or as E NC core and E ligand are increased. The band gap width behaves non-monotonically with d , L , E NC core , and E ligand , and intercoupling of these variables can eliminate the band gap. Lastly, we observe multiple phononic band gaps in many superlattices and find a correlation between an increase in the number of band gaps and increases in d and E NC core . We find that increases in the property mismatch between phononic crystal components ( i.e. d / L and E NC core / E ligand ) flattens the phonon branches and are a key driver in increasing the number of phononic band gaps. Our predicted phononic band gap center frequencies and widths far exceed those in current experimental demonstrations of 3-dimensional phononic crystals. This suggests that colloidal nanocrystal superlattices are promising candidates for use in high frequency phononic crystal applications.more » « less
-
Atomically precise nanoclusters (NCs) are of great interest due to their well-defined structures and molecule-like properties. Understanding their structure–property relationship is an important task because it can help tailor their structures to achieve specific desired properties. In this study, the temperature-dependent bonding properties of Ag44(SR)30 have been revealed by extended X-ray absorption fine structure (EXAFS) with a new structure analysis method, which includes two Ag–S and two Ag–Ag fitting shells. It has been proven that the EXAFS fitting quality can be improved significantly compared with the conventional method. New insights into Ag–S bondings were discovered based on the fitting results obtained from the new method. It allows us to observe two different bonding properties within the Ag–S motifs, which cannot be discovered by using the conventional method. Additionally, the metal core of Ag44(SR)30 exhibits uncommon thermal behavior, which could be connected to the absence of the center atom in the icosahedral core. Our results demonstrate that the new structure analysis method can provide a more reliable comparison of NCs structural changes than the conventional method, and it could be applicable to other NCs. The revealed temperature-dependent bonding properties can provide insights into the structure–property relationship of Ag44(SR)30, which can help design new NCs materials with tailored properties.more » « less
-
null (Ed.)ABSTRACT We investigate the time evolution of dense cores identified in molecular cloud simulations using dendrograms, which are a common tool to identify hierarchical structure in simulations and observations of star formation. We develop an algorithm to link dendrogram structures through time using the three-dimensional density field from magnetohydrodynamical simulations, thus creating histories for all dense cores in the domain. We find that the population-wide distributions of core properties are relatively invariant in time, and quantities like the core mass function match with observations. Despite this consistency, an individual core may undergo large (>40 per cent), stochastic variations due to the redefinition of the dendrogram structure between time-steps. This variation occurs independent of environment and stellar content. We identify a population of short-lived (<200 kyr) overdensities masquerading as dense cores that may comprise $$\sim\!20$$ per cent of any time snapshot. Finally, we note the importance of considering the full history of cores when interpreting the origin of the initial mass function; we find that, especially for systems containing multiple stars, the core mass defined by a dendrogram leaf in a snapshot is typically less than the final system stellar mass. This work reinforces that there is no time-stable density contour that defines a star-forming core. The dendrogram itself can induce significant structure variation between time-steps due to small changes in the density field. Thus, one must use caution when comparing dendrograms of regions with different ages or environment properties because differences in dendrogram structure may not come solely from the physical evolution of dense cores.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1103/PhysRevE.108.024311
