查看更多>>摘要：This study proposes an approach to the construction of granular models directly based on information granules expressed both in input and output spaces. Associating these information granules, the constructed granular models come in the framework of three layers networks: input granules, an inference scheme and output granules. The proposed approach consists of two stages. First, an augmented principle of justifiable granularity is proposed and applied to construct information granules in an input space. This principle constructs information granules not only through establishing a sound balance between two criteria, i.e., coverage and specificity, but also by optimizing those information granules on the basis of their homogeneity assessed with respect to data localized in output space. At the second stage, we propose an inference scheme by analyzing a location of an input datum in relation with the already formed information granules in an input space. The computed relation can be quantified as membership grades, thus yielding aggregation results involving information granules in an output space. The performance of the proposed granular model is supported by the mechanisms of granular computing and the principle of justifiable granularity. Experimental studies concerning synthetic and publicly available data are performed and some comparative analysis involving rule-based models is given. (C) 2021 Published by Elsevier B.V.
查看更多>>摘要：Mechanisms towards the automatic analog integrated circuit layout design have been an intensive research topic in the past few decades. Still, the industrial environment has no automatic approach established. The advances of machine learning applications in electronic design automation come with the promise to change this reality. This paper proposes a deep learning generative model for the placement "optimization"of analog integrated circuit basic blocks. The model behaves as an argmin operator for the placement cost function and can provide placement solutions instantly. Moreover, the model can be fed with unlabeled data, greatly facilitating data collection. A generic and innovative circuits' representation at the network's input layer is proposed, encoding the devices' dimensions, connectivity, and topological constraints. Besides, the randomness found in generative models is embedded directly into the feature vector, as the order of the features per device is shuffled in the input vector. Shuffling the order of the devices' features in the input not only brings multi-modality but also solves a generalization problem, as there is not any natural order defined to place devices in the feature vector. As a proof of concept, a deep artificial neural network capable of proposing different placement solutions, in less than 150 ms each, for six amplifier topologies and, in multiple technology nodes ranging from 350 nm down to 65 nm, is demonstrated. DeepPlacer was capable of producing correct solutions for topologies and technology nodes not present in the training set, showing good generalization while not hindering circuit performance due to the placement. (c) 2021 Elsevier B.V. All rights reserved.
查看更多>>摘要：In this paper, a novel ensemble model of one-dimensional convolution neural network (1DCNN) and support vector machine (SVM) is proposed to improve the detection accuracy in the process of pipeline leakage detection. Firstly, 1DCNN is constructed by experiments on different network structures and parameters, and it is used to extract data features adaptively. Then, an improved particle swarm optimization (PSO) algorithm is put forward, called variable amplitude PSO (VAPSO), with the adjustment strategy of parameter variable amplitude vibration to optimize the parameter combination in SVM and decrease the risk of trapping into local optimum in the training process. Finally, the data features extracted adaptively from the network are input into the improved VAPSO-SVM to classify. It is demonstrated by the experimental results that, compared with the existing models, the developed ensemble model has the capacity to extract the features of pipeline data more quickly and accurately with effective improvement in the classification accuracy, and has better robustness in the process of pipeline leakage detection. (C) 2021 Published by Elsevier B.V.
查看更多>>摘要：In many practical applications, such as fraud detection, credit risk modeling or medical decision making, classification models for assigning instances to a predefined set of classes are required to be both precise and interpretable. Linear modeling methods such as logistic regression are often adopted since they offer an acceptable balance between precision and interpretability. Linear methods, however, are not well equipped to handle categorical predictors with high cardinality or to exploit nonlinear relations in the data. As a solution, data preprocessing methods such as weight of evidence are typically used for transforming the predictors. The binning procedure that underlies the weight-of-evidence approach, however, has been little researched and typically relies on ad hoc or expert-driven procedures. The objective in this paper, therefore, is to propose a formalized, data-driven and powerful method. To this end, we explore the discretization of continuous variables through the binning of spline functions, which allows for capturing nonlinear effects in predictor variables and yields highly interpretable predictors that take only a small number of discrete values. Moreover, we extend the weight-of-evidence approach and propose to estimate the proportions using shrinkage estimators. Together, this method offers an improved ability to exploit both nonlinear and categorical predictors to achieve increased classification precision while maintaining the interpretability of the resulting model and decreasing the risk of overfitting. We present the results of a series of experiments in fraud detection and credit risk settings, which illustrate the effectiveness of the presented approach. (C) 2021 Elsevier B.V. All rights reserved.
NSTL
Elsevier