Tabular data: Deep learning is not all you need

A key element in solving real-life data science problems is selecting the types of models to use. Tree ensemble models (such as XGBoost) are usually recommended for classification and regression problems with tabular data. However, several deep learning models for tabular data have recently been proposed, claiming to outperform XGBoost for some use cases. This paper explores whether these deep models should be a recommended option for tabular data by rigorously comparing the new deep models to XGBoost on various datasets. In addition to systematically comparing their performance, we consider the tuning and computation they require. Our study shows that XGBoost outperforms these deep models across the datasets, including the datasets used in the papers that proposed the deep models. We also demonstrate that XGBoost requires much less tuning. On the positive side, we show that an ensemble of deep models and XGBoost performs better on these datasets than XGBoost alone.     

Efficient Flexible M-Tree Bulk Loading Using FastMap and Space-Filling Curves

Many database applications currently deal with objects in a metric space. Examples of such objects include unstructured multimedia objects and points of interest (POIs) in a road network. The M-tree is a dynamic index structure that facilitates an efficient search for objects in a metric space. Studies have been conducted on the bulk loading of large datasets in an M-tree. However, because previous algorithms involve excessive distance computations and disk accesses, they perform poorly in terms of their index construction and search capability. This study proposes two efficient M-tree bulk loading algorithms. Our algorithms minimize the number of distance computations and disk accesses using FastMap and a space-filling curve, thereby significantly improving the index construction and search performance. Our second algorithm is an extension of the first, and it incorporates a partitioning clustering technique and flexible node architecture to further improve the search performance. Through the use of various synthetic and real-world datasets, the experimental results demonstrated that our algorithms improved the index construction performance by up to three orders of magnitude and the search performance by up to 20.3 times over the previous algorithm.     

Effects on the crystallite growth behavior of LaPO4 nanopowders and its mechanical properties

The effects of pH of the reaction solution and the concentration of phosphoric acid on the crystal growth behavior of LaPO₄ crystallites were investigated and the mechanical properties of rare-earth phosphates were compared. As a result, the concentration of phosphoric acid of 10% was beneficial to the crystal growth of LaPO₄ nanocrystalline. When the pH value of the reaction solution was 2, the size of LaPO₄ crystallites increased gradually with the increasing reaction temperature, and the smallest crystallite size of 43.27 nm was obtained after heat-treatment at 1000 °C. Simultaneously, the activation energy for crystal growth of LaPO₄ nanocrystalline was relatively lower (26.82 kJ mol⁻¹). With the decreasing radii of rare-earth ions, the hardness, Young's modulus and fracture toughness of the bulk rare-earth phosphates exhibited a reduced tendency, resulted from the increase of porosity under the same preparation process.     

Epitaxial Growth of 2D Materials on High-Index Substrate Surfaces

Recently, the successful synthesis of wafer-scale single-crystal graphene, hexagonal boron nitride (hBN), and MoS₂ on transition metal surfaces with step edges boosted the research interests in synthesizing wafer-scale 2D single crystals on high-index substrate surfaces. Here, using hBN growth on high-index Cu surfaces as an example, a systematic theoretical study to understand the epitaxial growth of 2D materials on various high-index surfaces is performed. It is revealed that hBN orientation on a high-index surface is highly dependent on the alignment of the step edges of the surface as well as the surface roughness. On an ideal high-index surface, well-aligned hBN islands can be easily achieved, whereas curved step edges on a rough surface can lead to the alignment of hBN along with different directions. This study shows that high-index surfaces with a large step density are robust for templating the epitaxial growth of 2D single crystals due to their large tolerance for surface roughness and provides a general guideline for the epitaxial growth of various 2D single crystals.     

Biocompatibility and electron microscopy studies of epitaxial nanolaminate (Al0·5Ti0.5)N/ZrN coatings deposited by Arc-PVD technique

The ability to combine layers with high mechanical strength and additional physicochemical properties, such as biocompatibility, makes the use of multilayer coatings attractive for various applications. The transition from single layer to nanolaminate architecture can improve the mechanical performance of the coatings by increasing the number of interfaces and decreasing the modulation period of the layers. The microstructural study of the nanolaminate (Al_0·5Ti_0.5)N/ZrN coating with a modulation period λ of ≃ 20 nm was carried out using the TEM-HRTEM method. It was found that the coatings of (Al_0·5Ti_0.5)N/ZrN series consisted of two phases: the fcc-(Ti,Al)N solid solution obtained by isomorphic substitution of Ti atoms with Al ones in the TiN crystal lattice and the cubic ZrN phase. ZrN layers had a high texture structure with [111]-preferred growth texture and made a dominant effect on the nucleation and growth of (Al_0·5Ti_0.5)N layers. The epitaxial growth process was the most pronounced for fcc-(Ti,Al)N (111)||fcc-(ZrN) (111) and fcc-(Ti,Al)N (200)||fcc-(ZrN) (200) grains. Finally, the new coating demonstrated high biocompatibility, failure to toxicity and supported U2OS osteogenic cells proliferation within 7 days of cultivation.     

A thermal perspective of flash sintering: The effect of AC current ramp rate on microstructure evolution

In flash sintering experiments, the thermal history of the sample is key to understanding the mechanisms underlying densification rate and final properties. By combining robust temperature measurements with current-ramp-rate control, this study examined the effects of the thermal profile on the flash sintering of yttria-stabilized zirconia, with experiments ranging from a few seconds to several hours. The final density was maximized at slower heating rates, although processes slower than a certain threshold led to grain growth. The amount of grain growth observed was comparable to a similar conventional thermal process. The bulk electrical conductivity correlated with the maximum temperature and cooling rate. The only property that exhibited behavior that could not be attributed to solely the thermal profile was the grain boundary conductivity, which was consistently higher than conventional in flash sintered samples. These results suggest that, during flash sintering, athermal electric field effects are relegated to the grain boundary.     

Marine Phospholipids from Fishery By-Products Attenuate Atherosclerosis

In this study, the anti-atherosclerotic properties of three marine phospholipids (MPLs) extracts from fishery by-products including codfish roe, squid gonad, and shrimp head are verified. Their effects on key factors involved in atherosclerosis are examined and compared to explore whether the differences in their constitutions lead to the differences in the function. All three MPLs dampen oxidation of low- density lipoproteins (LDL) in vitro. Treating RAW264.7 macrophages and HUVECs endothelial cells with each MPLs ranging 10–100 µg mL⁻¹ does not decrease cell viability, yet ox-LDL caused cytotoxicity of both cells are alleviated by 50 or 100 µg mL⁻¹ MPLs treatment. In addition, the three MPLs reduce ox-LDL induced macrophage foam-like transition, mainly through inhibition of lipid uptake. Of the three MPLs, the one from squid gonad exhibits the best effect. On the other hand, all three MPLs modulate inflammatory responses, equally, by inhibiting the adhesion of monocytes to endothelial cells, and decreasing secretion of pro-inflammatory cytokines IL-6 and MCP-1. Using a high-cholesterol diet induced zebrafish model, it is found that all three MPLs, especially the one from squid gonad, alleviates cholesterol accumulation in early plaques, and decreases total cholesterol as well as lipid peroxide in vivo. Practical Applications: As a way of making the best of the increasingly scarce marine resources, valuable lipid components can be recovered from by-products and wastes from the fishery industry. Here, we tested the anti-atherosclerotic effects and the mechanisms of three MPLs extracted from codfish roe, squid gonad, and shrimp head. Our study provides further evidence that marine phospholipids extracted from fishery by-products could protect against atherosclerosis, and helps to elucidate the structure-function relationship of MPLs.     

Nitrogen-doped cobalt sulfide as an efficient electrocatalyst for hydrogen evolution reaction in alkaline and acidic media

The enhancement in intrinsic catalytic activity and material conductivity of an electrocatalyst can leads to promoting HER activity. Herein, a successful nitrogenation of CoS₂ (N–CoS₂) catalyst has been investigated through the facile hydrothermal process followed by N₂ annealing treatment. An optimized N–CoS₂ catalyst reveals an outstanding hydrogen evolution reaction (HER) performance in alkaline as well as acidic electrolyte media, exhibiting an infinitesimal overpotential of ?0.137 and ?0.097 V at a current density of ?10 mA/cm² (?0.309 and ?0.275 V at ?300 mA/cm²), corresponding respectively, with a modest Tafel slope of 117 and 101 mV/dec. Moreover, a static voltage response was observed at low and high current rates (?10 to ?100 mA/cm²) along with an excellent endurance up to 50 h even at ?100 mA/cm². The excellent catalytic HER performance is ascribed to improved electronic conductivity and enhanced electrochemically active sites, which is aroused from the synergy and mutual interaction between heteroatoms that might have varied the surface chemistry of an active catalyst.     

Grain orientation evolution and multi-scale interfaces enhanced thermoelectric properties of textured Sr0.9La0.1TiO3 based ceramics

Sr_0.9La_0.1TiO₃ based textured ceramics (SLTT-S3T) with a texture fraction of 0.81 are successfully fabricated by the reactive template grain growth method, in which Sr_0.9La_0.1TiO₃/20 wt%Ti was used as matrix and 10 wt% plate-like Sr₃Ti₂O₇ template seeds were used as templates. The phase transition, microstructure evolution, and the anisotropic thermoelectric properties of SLTT-S3T ceramics were investigated. The results show that the ceramics are mainly composed of Sr_0.9La_0.1TiO₃ and rutile TiO₂ phases. Grains grow with a preferred orientation along (h00). A maximum ZT of 0.26 at 1073 K was achieved in the direction perpendicular to the tape casting direction. The low lattice thermal conductivity of 1.9 W/(m K) at 1073 K was obtained decreased by 34%, 40%, and 38% compared with non-textured, SrTiO₃ and Sr_0.9La_0.1TiO₃ ceramics prepared by the same process, can be attributed to the enhanced phonon scattering by the complex multi-scale boundaries and interfaces. This work provides a strategy of microstructural design for thermoelectric oxides to decrease intrinsic lattice thermal conductivity and further regulate thermoelectric properties via texture engineering.