Search for: All records

Context.Traditionally, supersonic turbulence is considered to be one of the most likely mechanisms slowing the gravitational collapse in dense clumps, thereby enabling the formation of massive stars. However, several recent studies have raised differing points of view based on observations carried out with sufficiently high spatial and spectral resolution. These studies call for a re-evaluation of the role turbulence plays in massive star-forming regions. Aims.Our aim is to study the gas properties, especially the turbulence, in a sample of massive star-forming regions with sufficient spatial and spectral resolution, which can both resolve the core fragmentation and the thermal line width. Methods.We observed NH₃metastable lines with the Very Large Array (VLA) to assess the intrinsic turbulence. Results.Analysis of the turbulence distribution histogram for 32 identified NH₃cores reveals the presence of three distinct components. Furthermore, our results suggest that (1) sub- and transonic turbulence is a prevalent (21 of 32) feature of massive star-forming regions and those cold regions are at early evolutionary stage. This investigation indicates that turbulence alone is insufficient to provide the necessary internal pressure required for massive star formation, necessitating further exploration of alternative candidates; and (2) studies of seven multi-core systems indicate that the cores within each system mainly share similar gas properties and masses. However, two of the systems are characterized by the presence of exceptionally cold and dense cores that are situated at the spatial center of each system. Our findings support the hub-filament model as an explanation for this observed distribution. 

Learning fair representations is an essential task to reduce bias in data-oriented decision making. It protects minority subgroups by requiring the learned representations to be independent of sensitive attributes. To achieve independence, the vast majority of the existing work primarily relaxes it to the minimization of the mutual information between sensitive attributes and learned representations. However, direct computation of mutual information is computationally intractable, and various upper bounds currently used either are still intractable or contradict the utility of the learned representations. In this paper, we introduce distance covariance as a new dependence measure into fair representation learning. By observing that sensitive attributes (e.g., gender, race, and age group) are typically categorical, the distance covariance can be converted to a tractable penalty term without contradicting the utility desideratum. Based on the tractable penalty, we propose FairDisCo, a variational method to learn fair representations. Experiments demonstrate that FairDisCo outperforms existing competitors for fair representation learning. 

Co-evolving sequences are ubiquitous in a variety of applications, where different sequences are often inherently inter-connected with each other. We refer to such sequences, together with their inherent connections modeled as a structured network, as network of co-evolving sequences (NoCES). Typical NoCES applications in- clude road traffic monitoring, company revenue prediction, motion capture, etc. To date, it remains a daunting challenge to accurately model NoCES due to the coupling between network structure and sequences. In this paper, we propose to modeling NoCES with the aim of simultaneously capturing both the dynamics and the inter- play between network structure and sequences. Specifically, we propose a joint learning framework to alternatively update the network representations and sequence representations as the se- quences evolve over time. A unique feature of our framework lies in that it can deal with the case when there are co-evolving sequences on both network nodes and edges. Experimental evaluations on four real datasets demonstrate that the proposed approach (1) out- performs the existing competitors in terms of prediction accuracy, and (2) scales linearly w.r.t. the sequence length and the network size. 

Hybrid microfluidic systems that are composed of multiple different types of substrates have been recognized as a versatile and superior platform, which can draw benefits from different substrates while avoiding their limitations. This review article introduces the recent innovations of different types of low-cost hybrid microfluidic devices, particularly focusing on cost-effective polymer- and paper-based hybrid microfluidic devices. In this article, the fabrication of these hybrid microfluidic devices is briefly described and summarized. We then highlight various hybrid microfluidic systems, including polydimethylsiloxane (PDMS)-based, thermoplastic-based, paper/polymer hybrid systems, as well as other emerging hybrid systems (such as thread-based). The special benefits of using these hybrid systems have been summarized accordingly. A broad range of biological and biomedical applications using these hybrid microfluidic devices are discussed in detail, including nucleic acid analysis, protein analysis, cellular analysis, 3D cell culture, organ-on-a-chip, and tissue engineering. The perspective trends of hybrid microfluidic systems involving the improvement of fabrication techniques and broader applications are also discussed at the end of the review. 

Attributed network embedding aims to learn low dimensional node representations by combining both the network's topological structure and node attributes. Most of the existing methods either propagate the attributes over the network structure or learn the node representations by an encoder-decoder framework. However, propagation based methods tend to prefer network structure to node attributes, whereas encoder-decoder methods tend to ignore the longer connections beyond the immediate neighbors. In order to address these limitations while enjoying the best of the two worlds, we design cross fusion layers for unsupervised attributed network embedding. Specifically, we first construct two separate views to handle network structure and node attributes, and then design cross fusion layers to allow flexible information exchange and integration between the two views. The key design goals of the cross fusion layers are three-fold: 1) allowing critical information to be propagated along the network structure, 2) encoding the heterogeneity in the local neighborhood of each node during propagation, and 3) incorporating an additional node attribute channel so that the attribute information will not be overshadowed by the structure view. Extensive experiments on three datasets and three downstream tasks demonstrate the effectiveness of the proposed method. 

Abstract One of the most poorly understood aspects of low-mass star formation is how multiple-star systems are formed. Here we present the results of Atacama Large Millimeter/submillimeter Array (ALMA) Band 6 observations toward a forming quadruple protostellar system, G206.93-16.61E2, in the Orion B molecular cloud. ALMA 1.3 mm continuum emission reveals four compact objects, of which two are Class I young stellar objects and the other two are likely in prestellar phase. The 1.3 mm continuum emission also shows three asymmetric ribbon-like structures that are connected to the four objects, with lengths ranging from ∼500 to ∼2200 au. By comparing our data with magnetohydrodynamic simulations, we suggest that these ribbons trace accretion flows and also function as gas bridges connecting the member protostars. Additionally, ALMA CO J = 2−1 line emission reveals a complicated molecular outflow associated with G206.93-16.61E2, with arc-like structures suggestive of an outflow cavity viewed pole-on.