Although the deep learning technique has been introduced into computational physics in recent years, feasibility of applying it to solve electromagnetic (EM) scattering field from arbitrary scatters remains open. In this article, convolutional neural network (CNN) employed predict EM scattered by complex geometries under plane-wave illumination. The 2-D finite-difference frequency-domain (FDFD) algorithm, wrapped a module randomly generate basic geometries, is produce training data for...

Owning to the nature of flood events, near-real-time detection and mapping is essential for disaster prevention, relief, mitigation. In recent years, rapid advancement deep learning has brought endless possibilities field detection. However, relies heavily on training samples availability high-quality datasets rather limited. The present study collected 16 events in Yangtze River Basin divided them into three categories different purpose: training, testing, application. An efficient...

The task of partial label (PL) learning is to learn a multi-class classifier from training examples each associated with set candidate labels, among which only one corresponds the ground-truth label. It well known that for inducing predictive model, most straightforward solution binary decomposition works by either one-vs-rest or one-vs-one strategy. Nonetheless, PL example concealed in its and thus not accessible algorithm, cannot be directly applied under scenario. In this paper, novel...

Multi-view multi-label learning serves an important framework to learn from objects with diverse representations and rich semantics. Existing multi-view techniques focus on exploiting shared subspace for fusing representations, where helpful view-specific information discriminative modeling is usually ignored. In this paper, a novel approach named SIMM proposed which leverages exploitation extraction. For exploitation, jointly minimizes confusion adversarial loss utilize all views....

Identifying and assessing the disaster risk of landslide-prone regions is very critical for prevention mitigation. Owning to their special advantages, neural network algorithms have been widely used landslide susceptibility mapping (LSM) in recent decades. In present study, three advanced models popularly relevant studies, i.e. artificial (ANN), one dimensional convolutional (1D CNN) recurrent (RNN), were evaluated compared LSM practice over Qingchuan County, Sichuan province, China....

Canonical correlation analysis (CCA) is an effective spatial filtering algorithm widely used in steady-state visual evoked potential (SSVEP)-based brain-computer interfaces (BCIs). In existing CCA methods, training data are for constructing templates of stimulus targets and the filters created between template signals a single-trial testing signal. The fact that rely on data, however, results low classification performance compared to other state-of-the-art algorithms such as task-related...

In multi-view multi-label learning (MVML), each training example is represented by different feature vectors and associated with multiple labels simultaneously. Nonetheless, the labeling quality of examples tend to be affected annotation noises. this paper, problem partial (MVPML) studied, where set are assumed candidate ones only partially valid. To solve MVPML problem, a two-stage graph-based disambiguation approach proposed. Firstly, ground-truth estimated disambiguating fused similarity...

Natural human-robot interaction requires different and more robust models of language understanding (NLU) than non-embodied NLU systems. In particular, architectures are required that (1) process incrementally in order to be able provide early backchannel feedback human speakers; (2) use pragmatic contexts throughout the infer missing information; (3) handle underspecified, fragmentary, or otherwise ungrammatical utterances common spontaneous speech. this paper, we describe our attempts at...

We study the problem of constructing ε-coresets for (k, z)-clustering in a doubling metric M(X, d). An ε-coreset is weighted subset S ⊆ X with weight function w : → ℝ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">≥0</sub> , such that any k-subset C ∈ [X] <sup xmlns:xlink="http://www.w3.org/1999/xlink">k</sup> it holds Σ xmlns:xlink="http://www.w3.org/1999/xlink">x∈S</sub> w(x) · d xmlns:xlink="http://www.w3.org/1999/xlink">z</sup> (x, C) (1 ±...

Blue and Green Infrastructures (BGIs) are natural or semi-natural systems that considered an efficient solution to enhance stormwater management. To assess the performance of BGIs in mitigating floods droughts urban area, a water balance model was developed this study simulate runoff formation propagation. The features fine spatial temporal resolutions flexibly integrates BGIs. Combining conceptual single reservoir approach empirical continuous Soil Conservation Service Curve Number (SCS-CN)...

The identification and early warning of potential landslides can effectively reduce the number casualties amount property loss. At present, interferometric synthetic aperture radar (InSAR) is considered one mainstream methods for large-scale detection landslides, it obtain long-term time-series surface deformation data. However, method identifying anomalous areas using InSAR data still mainly manual delineation, which time-consuming, labor-consuming, has no generally accepted criterion. In...

In conventional multi-label learning framework, each example is assumed to be represented by a single feature vector and associated with multiple valid labels simultaneously. Nonetheless, real-world objects usually exhibit complicated properties which can have multi-view representation as well false positive labeling. Accordingly, the problem of partial (MVPML) studied in this paper, where presented vectors while candidate are only partially valid. To learn from MVPML examples, novel...

With the explosive growth in demand for lithium (Li) resources, Mufushan area has been a hotspot Li deposit exploration China recent years. Geochemical maps and geochemical anomaly are basic of mineral resources. A fixed-value method to contour map is presented here, which concentrations divided into 19 levels on 18 fixed values, ranging from 5 μg/g (corresponding detection limit) 1858 cut-off grade hard-rock type) illustrated six color tones corresponding areas low background, high anomaly,...

Semi-supervised learning (SSL) is effectively used for numerous classification problems, thanks to its ability make use of abundant unlabeled data. The main assumption various SSL algorithms that the nearby points on data manifold are likely share a label. Graph-based constructs graph from point-cloud as an approximation underlying manifold, followed by label inference. It no surprise quality constructed in capturing essential structure critical accuracy subsequent inference step [6].

Geochemical gene is a new promising concept proposed recently in the discrimination and traceability of geological materials also useful tool to recognize geochemical anomalies mineral exploration. Based on lithogenes LG01 LG03, can be classified into nine types LG_CR compositionally. With respect with 11 LG_CR, order eliminate lithological influence further narrow prospecting target area, tungsten metallogene named MGW11 for exploration after MGW. Six weathering profiles developed granitic...

Convolutional neural network (CNN) is capable of automatically extracting image features and has been widely used in remote sensing classifications. Feature extraction an important difficult problem current research. In this paper, data augmentation for avoiding over fitting was attempted to enrich samples improve the performance a newly proposed convolutional with UC-Merced RSI-CB datasets remotely sensed scene A multiple grouped (MGCNN) self-learning that promoting efficiency CNN proposed,...

Latest Permian to Triassic plutons are widespread in the northern North China Craton (NCC); most of them show calc-alkaline, high-K or alkaline geochemical features. The Shadegai pluton Wulashan area has shoshonitic affinity and I-type character, is composed syenogranites containing abundant mafic microgranular enclaves (MMEs). LA-MC-ICP-MS U–Pb data yield weighted mean 206Pb/238U ages 222 ± 1 Ma 221 for MMEs, respectively, suggesting their coeval formation during Late Triassic. have high...