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.     

Learn decision trees with deep visual primitives

In this paper, we strive to propose a self-interpretable framework, termed PrimitiveTree, that incorporates deep visual primitives condensed from deep features with a conventional decision tree, bridging the gap between deep features extracted from deep neural networks (DNNs) and trees’ transparent decision-making processes. Specifically, we utilize a codebook, which embeds the continuous deep features into a finite discrete space (deep visual primitives) to distill the most common semantic information. The decision tree adopts the spatial location information and the mapped primitives to present the decision-making process of the deep features in a tree hierarchy. Moreover, the trained interpretable PrimitiveTree can inversely explain the constituents of the deep features, highlighting the most critical and semantic-rich image patches attributing to the final predictions of the given DNN. Extensive experiments and visualization results validate the effectiveness and interpretability of our method.     

Enhanced wide-activated residual network for efficient and accurate image deblocking

To save bandwidth and storage space as well as speed up data transmission, people usually perform lossy compression on images. Although the JPEG standard is a simple and effective compression method, it usually introduces various visually unpleasing artifacts, especially the notorious blocking artifacts. In recent years, deep convolutional neural networks (CNNs) have seen remarkable development in compression artifacts reduction. Despite the excellent performance, most deep CNNs suffer from heavy computation due to very deep and wide architectures. In this paper, we propose an enhanced wide-activated residual network (EWARN) for efficient and accurate image deblocking. Specifically, we propose an enhanced wide-activated residual block (EWARB) as basic construction module. Our EWARB gives rise to larger activation width, better use of interdependencies among channels, and more informative and discriminative non-linearity activation features without more parameters than residual block (RB) and wide-activated residual block (WARB). Furthermore, we introduce an overlapping patches extraction and combination (OPEC) strategy into our network in a full convolution way, leading to large receptive field, enforced compatibility among adjacent blocks, and efficient deblocking. Extensive experiments demonstrate that our EWARN outperforms several state-of-the-art methods quantitatively and qualitatively with relatively small model size and less running time, achieving a good trade-off between performance and complexity.     

Generating Questions Based on Semi-Automated and End-to-End Neural Network

With the emergence of large-scale knowledge base, how to use triple information to generate natural questions is a key technology in question answering systems. The traditional way of generating questions require a lot of manual intervention and produce lots of noise. To solve these problems, we propose a joint model based on semi-automated model and End-to-End neural network to automatically generate questions. The semi-automated model can generate question templates and real questions combining the knowledge base and center graph. The End-to-End neural network directly sends the knowledge base and real questions to BiLSTM network. Meanwhile, the attention mechanism is utilized in the decoding layer, which makes the triples and generated questions more relevant. Finally, the experimental results on SimpleQuestions demonstrate the effectiveness of the proposed approach.     

基于神经网络和遗传算法的锭子弹性管性能优化

为得到减振弹性管对下锭胆的支承弹性和锭子高速运动下的稳定性等性能的最优匹配效率,依据减振弹性管的等效抗弯刚度及底部等效刚度系数公式,利用MatLab数值分析软件构建弹性管抗弯刚度和底部挠度数学模型。首先,结合Isight优化软件基于径向基神经网络构建其近似模型,且使精度达到可接受水平,并以模型的关键结构参数弹性模量、螺距、槽宽、壁厚为设计变量,结合遗传算法对弹性管抗弯刚度和底部挠度进行多目标优化设计,得到Pareto最优解集和Pareto前沿图,确定出减振弹性管结构工艺参数的优化方案。通过对优化数据进行分析发现,该方案在保证减振弹性管弹性的同时,其底部振幅明显减弱。     

Classification of the cardiotocogram data for anticipation of fetal risks using machine learning techniques

The aim of the research is evaluating the classification performances of eight different machine-learning methods on the antepartum cardiotocography (CTG) data. The classification is necessary to predict newborn health, especially for the critical cases. Cardiotocography is used for assisting the obstetricians’ to obtain detailed information during the pregnancy as a technique of measuring fetal well-being, essentially in pregnant women having potential complications. The obstetricians describe CTG shortly as a continuous electronic record of the baby's heart rate took from the mother's abdomen. The acquired information is necessary to visualize unhealthiness of the embryo and gives an opportunity for early intervention prior to happening a permanent impairment to the embryo. The aim of the machine learning methods is by using attributes of data obtained from the uterine contraction (UC) and fetal heart rate (FHR) signals to classify as pathological or normal. The dataset contains 1831 instances with 21 attributes, examined by applying the methods. In the paper, the highest accuracy displayed as 99.2%.     

Comparative evaluation of different statistical tools for the prediction of uniaxial compressive strength of rocks

In this study, uniaxial compressive strength (UCS), unit weight (UW), Brazilian tensile strength (BTS), Schmidt hardness (SHH), Shore hardness (SSH), point load index (Is₅₀) and P-wave velocity (V_p) properties were determined. To predict the UCS, simple regression (SRA), multiple regression (MRA), artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS) and genetic expression programming (GEP) have been utilized. The obtained UCS values were compared with the actual UCS values with the help of various graphs. Datasets were modeled using different methods and compared with each other. In the study where the performance indice PI_at was used to determine the best performing method, MRA method is the most successful method with a small difference. It is concluded that the mean PI_at equal to 2.46 for testing dataset suggests the superiority of the MRA, while these values are 2.44, 2.33, and 2.22 for GEP, ANFIS, and ANN techniques, respectively. The results pointed out that the MRA can be used for predicting UCS of rocks with higher capacity in comparison with others. According to the performance index assessment, the weakest model among the nine model is P7, while the most successful models are P2, P9, and P8, respectively.     

A bat-neural network multi-agent system (BNNMAS) for stock price prediction: Case study of DAX stock price

Creating an intelligent system that can accurately predict stock price in a robust way has always been a subject of great interest for many investors and financial analysts. Predicting future trends of financial markets is more remarkable these days especially after the recent global financial crisis. So traders who access to a powerful engine for extracting helpful information throw raw data can meet the success. In this paper we propose a new intelligent model in a multi-agent framework called bat-neural network multi-agent system (BNNMAS) to predict stock price. The model performs in a four layer multi-agent framework to predict eight years of DAX stock price in quarterly periods. The capability of BNNMAS is evaluated by applying both on fundamental and technical DAX stock price data and comparing the outcomes with the results of other methods such as genetic algorithm neural network (GANN) and some standard models like generalized regression neural network (GRNN), etc. The model tested for predicting DAX stock price a period of time that global financial crisis was faced to economics. The results show that BNNMAS significantly performs accurate and reliable, so it can be considered as a suitable tool for predicting stock price specially in a long term periods.     

Failed Neuroprotection of Combined Inhibition of L-Type and ASIC1a Calcium Channels with Nimodipine and Amiloride

Effective pharmacological neuroprotection is one of the most desired aims in modern medicine. We postulated that a combination of two clinically used drugs—nimodipine (L-Type voltage-gated calcium channel blocker) and amiloride (acid-sensing ion channel inhibitor)—might act synergistically in an experimental model of ischaemia, targeting the intracellular rise in calcium as a pathway in neuronal cell death. We used organotypic hippocampal slices of mice pups and a well-established regimen of oxygen-glucose deprivation (OGD) to assess a possible neuroprotective effect. Neither nimodipine (at 10 or 20 µM) alone or in combination with amiloride (at 100 µM) showed any amelioration. Dissolved at 2.0 Vol.% dimethyl-sulfoxide (DMSO), the combination of both components even increased cell damage (p = 0.0001), an effect not observed with amiloride alone. We conclude that neither amiloride nor nimodipine do offer neuroprotection in an in vitro ischaemia model. On a technical note, the use of DMSO should be carefully evaluated in neuroprotective experiments, since it possibly alters cell damage.     

Specific emitter identification for satellite communication using probabilistic neural networks

Electromagnetic signal emitted by satellite communication (satcom) transmitters are used to identify specific individual uplink satcom terminals sharing the common transponder in real environment, which is known as specific emitter identification (SEI) that allows for early indications and warning (I&W) of the targets carrying satcom furnishment and furthermore the real time electromagnetic situation awareness in military operations. In this paper, the authors are the first to propose the identification of specific transmitters of satcom by using probabilistic neural networks (PNN) to reach the goal of target recognition. We have been devoted to the examination by exploring the feasibility of utilizing the Hilbert transform to signal preprocessing, applying the discrete wavelet transform to feature extraction, and employing the PNN to perform the classification of stationary signals. There are a total of 1000 sampling time series with binary phase shift keying (BPSK) modulation originated by five types of satcom transmitters in the test. The established PNNs classifier implements the data testing and finally yields satisfactory accuracy at 8 dB(±1 dB) carrier to noise ratio, which indicates the feasibility of our method, and even the keen insight of its application in military.