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.     

IRS辅助的SWIPT系统中基于能效优先的波束成形设计与优化

以智能反射面(intelligent reflecting surface,IRS)辅助的无线携能通信(simultaneous wireless information and power transfer,SWIPT)系统为背景,研究了该系统中基于能效优先的多天线发送端有源波束成形与IRS无源波束成形联合设计与优化方法。以最大化接收端的最小能效为优化目标,构造在发送端功率、接收端能量阈值、IRS相移等多约束下的非线性优化问题,用交替方向乘子法(alternating direction method of multipliers,ADMM)求解。采用Dinkelbach算法转化目标函数,通过奇异值分解(singular value decomposition,SVD)和半定松弛(semi-definite relaxation,SDR)得到发送端有源波束成形向量。采用SDR得到IRS相移矩阵与反射波束成形向量。结果表明,该系统显著降低了系统能量收集(energy harvesting,EH)接收端的能量阈值。当系统总电路功耗为?15 dBm时,所提方案的用户能效为300 KB/J。当IRS反射阵源数与发送天线数均为最大值时,系统可达最大能效。     

Research and development of liquid hydrogen (LH2) temperature monitoring system for marine applications

Monitoring the temperature in liquid hydrogen (LH₂) storage tanks on ships is important for the safety of maritime navigation. In addition, accurate temperature measurement is also required for commercial transactions. Temperature and pressure define the density of liquid hydrogen, which is directly linked to trading interests. In this study, we developed and tested a liquid hydrogen temperature monitoring system that uses platinum resistance sensors with a nominal electrical resistance of approximately 1000 Ω at room temperature, PT-1000, for marine applications. The temperature measurements were carried out using a newly developed temperature monitoring system under different pressure conditions. The measured values are compared with a calibrated reference PT-1000 resistance thermometer. We confirm a measurement accuracy of ±50 mK in a pressure range of 0.1 MPa–0.5 MPa.     

Effect of heat treatment on structural,morphological, dielectric and magnetic properties of Mg–Zn ferrite nanoparticles

Green combustion was used to prepare a ferrite composition of Mg_0.4Zn_0.6Fe₂O₄ using a blend of fresh lemon juice as a natural fuel-reductant. Effect of heat treatment on phase, morphological, dielectric, and humidity sensor properties is discussed. The formation of a cubic spinel ferrite has been established by XRD-diffraction and vibrational spectroscopic studies. The experimental lattice parameter ranges from 8.3721 to 8.3631 Å. The broadening of octahedral band (υ₂) in the vibrational spectra is an identification for the existence of ferrite nanoparticles in various sizes. The typical crystallite size ranges from 10.2 to 36.9 nm. Using micrographs obtained from field-effect scanning electron microscopy (FESEM), researchers observed a spherical-shaped microstructure with agglomerated nanoparticles. Dielectric investigations have shown that the current ferrite composition has typical dielectric dispersion. The highest reported value for saturation magnetization (M_s) in the present study is 33 emu/g. Magnetic behaviour is primarily influenced by magnetocrystalline anisotropy, cation distribution, and crystallite size. The existence of void spaces in the sintered samples, as well as their porous nature, rendered them suitable for humidity sensor applications. Sintered samples have good sensing capability at 900 °C. The current findings are integrated in terms of cation distribution and magnetocrystalline anisotropy, assuming fine size effects of ferrite nanoparticles.     

Ammonia gas sensing and magnetic permeability of enhanced surface area and high porosity lanthanum substituted Co–Zn nano ferrites

This work reports magnetic permeability and ammonia gas sensing characteristics of La³⁺ substituted Co–Zn nano ferrites possessing chemical formula Co_0.7Zn_0.3La_xFe_2-2xO₄ (x = 0–0.1) synthesized by a sol-gel route. Refinement of X-ray diffraction (XRD) patterns of the ferrite powders by the Rietveld technique has revealed the creation of single-phase spinel structure. The tenancy of constituent cations at tetrahedral/octahedral sites was obtained from the refinement of XRD. The crystallite sizes calculated from the W–H method vary from 20 to 24 nm. The scanning electron microscope (SEM) profiles of the ferrite samples were analyzed for the morphological details. The energy dispersive X-ray analysis (EDAX) patterns of the samples were obtained to test the elemental purity of the ferrites within their stoichiometry. The transmission electron microscope (TEM) image of the ferrite (x = 0.1) exhibits the spherical and oval shaped particles with a mean size of 20 nm. Fourier transform infra-red (FTIR) spectra were analyzed to confirm the superseding of La³⁺ cations at octahedral sites. The Brunauer-Emmett-Teller (BET) analysis of nitrogen adsorption-desorption isotherms of the ferrites was performed to investigate the porous structure and to determine the surface area of the nanocrystalline ferrites. The oxidation states of the constituent ions were confirmed by means of X-ray photoelectron spectroscopy (XPS). The complex permeability as a function of frequency was studied to explore the effects of structural parameters on the magnetic behaviour of the ferrites. Analysis of gas sensing properties of the ferrites have proved that the Co–Zn–La ferrite with controlled La composition can be utilized as an effective ammonia gas sensing material in commercial gas sensors.     

Potential influence of microorganisms on the corrosion of carbon steel in the French high- and intermediate-level long-lived radioactive waste disposal context

In the context of the high-level radioactive waste disposal CIGEO, the corrosion rate due to microbially influenced corrosion (MIC) has to be evaluated. In France, it is envisaged to dispose of high- and intermediate-level long-lived radioactive waste at a depth of 500 m in a deep geological disposal, drilled in the Callovo-Oxfordian claystone (Cox) formation. To do so, a carbon steel casing will be inserted inside disposal cells, which are horizontal tunnels drilled in the Cox. A specific cement grout will be injected between the carbon steel casing and the claystone. A study was conducted to evaluate the possibility of MIC on carbon steel in the foreseeable high radioactive waste disposal. The corrosiveness of various environments was investigated at 50°C and 80°C with or without microorganisms enriched from samples of Andra's underground research laboratory. The monitoring of corrosion during the experiments was ensured using gravimetric method and real-time corrosion monitoring using sensors based on the measurements of the electrical resistance. The corrosion data were completed with microbiological analyses including cultural and molecular characterizations.     

Neural network-based flight control systems: Present and future

As the first review in this field, this paper presents an in-depth mathematical view of Intelligent Flight Control Systems (IFCSs), particularly those based on artificial neural networks. The rapid evolution of IFCSs in the last two decades in both the methodological and technical aspects necessitates a comprehensive view of them to better demonstrate the current stage and the crucial remaining steps towards developing a truly intelligent flight management unit. To this end, in this paper, we will provide a detailed mathematical view of Neural Network (NN)-based flight control systems and the challenging problems that still remain. The paper will cover both the model-based and model-free IFCSs. The model-based methods consist of the basic feedback error learning scheme, the pseudocontrol strategy, and the neural backstepping method. Besides, different approaches to analyze the closed-loop stability in IFCSs, their requirements, and their limitations will be discussed in detail. Various supplementary features, which can be integrated with a basic IFCS such as the fault-tolerance capability, the consideration of system constraints, and the combination of NNs with other robust and adaptive elements like disturbance observers, would be covered, as well. On the other hand, concerning model-free flight controllers, both the indirect and direct adaptive control systems including indirect adaptive control using NN-based system identification, the approximate dynamic programming using NN, and the reinforcement learning-based adaptive optimal control will be carefully addressed. Finally, by demonstrating a well-organized view of the current stage in the development of IFCSs, the challenging issues, which are critical to be addressed in the future, are thoroughly identified. As a result, this paper can be considered as a comprehensive road map for all researchers interested in the design and development of intelligent control systems, particularly in the field of aerospace applications.     

A Voltage Control Method for a Distributed Line Feeder by means of Multiagent Approach

This paper proposes a method for the coordinated control of power factor by means of a multiagent approach. The proposed multiagent system consists of two types of agent: single feeder agent (F_AG) and bus agent (B_AG). In the proposed system, an F_AG plays as an important role, which decides the power factors of all distributed generators by executing the load flow calculations repeatedly. The voltage control strategies are implemented as the class definition of Java into the system. In order to verify the performance of the proposed method, it has been applied to a typical distribution model system. The simulation results show that the system is able to control very violent fluctuation of the demands and the photovoltaic (PV) generations.     

A review of topic modeling methods

Topic modeling is a popular analytical tool for evaluating data. Numerous methods of topic modeling have been developed which consider many kinds of relationships and restrictions within datasets; however, these methods are not frequently employed. Instead many researchers gravitate to Latent Dirichlet Analysis, which although flexible and adaptive, is not always suited for modeling more complex data relationships. We present different topic modeling approaches capable of dealing with correlation between topics, the changes of topics over time, as well as the ability to handle short texts such as encountered in social media or sparse text data. We also briefly review the algorithms which are used to optimize and infer parameters in topic modeling, which is essential to producing meaningful results regardless of method. We believe this review will encourage more diversity when performing topic modeling and help determine what topic modeling method best suits the user needs.     

Deep reinforcement learning-based cooperative interactions among heterogeneous vehicular networks

Most real-world vehicle nodes can be structured into an interconnected network of vehicles. Through structuring these services and vehicle device interactions into multiple types, such internet of vehicles becomes multidimensional heterogeneous overlay networks. The heterogeneousness of the overlays makes it difficult for the overlay networks to coordinate with each other to improve their performance. Therefore, it poses an interesting but critical challenge to the effective analysis of heterogeneous virtual vehicular networks. A variety of virtual vehicular networks can be easily deployed onto the native network by applying the concept of SDN (Software Defined Networking). These virtual networks reflect their heterogeneousness due to their different performance goals, and they compete for the same physical resources of the underlying network, so that a sub-optimal performance of the virtual networks may be achieved. Therefore, we propose a Deep Reinforcement Learning (DRL) approach to make the virtual networks cooperate with each other through the SDN controller. A cooperative solution based on the asymmetric Nash bargaining is proposed for co-existing virtual networks to improve their performance. Moreover, the Markov Chain model and DRL resolution are introduced to leverage the heterogeneous performance goals of virtual networks. The implementation of the approach is introduced, and simulation results confirm the performance improvement of the latency sensitive, loss-rate sensitive and throughput sensitive heterogeneous vehicular networks using our cooperative solution.