Deep learning technologies have been successfully applied to hyperspectral(HS) image classification with remarkable performance. However, compared with traditional machine learning methods, neural ...networks usually need more data. In remote sensing(RS) research, obtaining a large number of labeled HS data is a very difficult and expensive work. Simultaneously, the distribution of feature information is bound to be unbalance, and tends to conform to the long tail. At present, the neighborhood information of unlabeled samples is usually ignored in HS image classification tasks based on semi-supervised learning. In this letter, we propose a new semi-supervised long tail learning framework based on spatial neighborhood information(SLN-SNI), which can complete the HS image classification task under unbalanced small sample data. Specifically, a new semi-supervised learning strategy is proposed. On this basis, a new method to determine the label of unlabeled samples based on spatial neighborhood information(SNI) is proposed. The coarse classification results divided into three situations are judged again, and the accuracy of pseudo labels is improved. The performance of the proposed method is tested on three public HS image datasets. Compared with the current advanced methods have achieved a certain improvement.
In recent years, it has been a research hotspot to apply big data-driven deep learning methods to synthetic aperture radar (SAR) target recognition with limited data. However, the problem caused by ...the long-tailed characteristics of SAR data has long been ignored. Specifically, a majority of data samples are concentrated in a few categories, leading to a skewed distribution of data. This skewed distribution can cause learning bias toward the majority class, which can subsequently degrade the recognition performance of the minority class. This issue is further exacerbated in limited sample conditions for SAR target recognition. After conducting research on target recognition for long-tailed natural images, this study has found that the existing methods used in this field cannot be easily applied to SAR target recognition. The primary reason is that SAR image data exhibit simultaneous and complex interclass and intraclass long-tailed distributions. In response to this issue, we propose the use of a multibranch expert network and dual-environment sampling to address the long-tail problems in both interclass and intraclass scenarios. The proposed method outperforms popular long-tailed target recognition methods on the long-tailed versions of the MSTAR and FUSAR datasets.
Large-scale facial expression datasets are primarily composed of real-world facial expressions. Expression occlusion and large-angle faces are two important problems affecting the accuracy of ...expression recognition. Moreover, because facial expression data in natural scenes commonly follow a long-tailed distribution, trained models tend to recognize the majority classes while recognizing the minority classes with low accuracies. To improve the robustness and accuracy of expression recognition networks in an uncontrolled environment, this paper proposes an efficient network structure based on an attention mechanism that fuses global and local features (AM-FGL). We use a channel spatial model and local feature convolutional neural networks to perceive the global and local features of the human face, respectively. Because the distribution of real-world scene field expression datasets commonly follows a long-tail distribution, where neutral and happy expressions account for the tail expressions, a trained model exhibits low recognition accuracy for tail expressions such as fear and disgust. CutMix is a novel data enhancement method proposed in other fields; thus, based on the CutMix concept, a simple and effective data-balancing method is proposed (BC-EDB). The key idea is to paste key pixels (around eyes, mouths, and noses), which reduces the influence of overfitting. Our proposed method is more focused on the recognition of tail expression, occluded expression, and large-angle faces, and we achieved the most advanced results in occlusion-RAF-DB, 30∘ pose-RAF-DB, and 45∘ pose-RAF-DB with accuracies of 86.96%, 89.74%, and 88.53%.
Cortical activity involves large populations of neurons, even when it is limited to functionally coherent areas. Electrophysiological recordings, on the other hand, involve comparatively small neural ...ensembles, even when modern-day techniques are used. Here we review results which have started to fill the gap between these two scales of inquiry, by shedding light on the statistical distributions of activity in large populations of cells. We put our main focus on data recorded in awake animals that perform simple decision-making tasks and consider statistical distributions of activity throughout cortex, across sensory, associative, and motor areas. We transversally review the complexity of these distributions, from distributions of firing rates and metrics of spike-train structure, through distributions of tuning to stimuli or actions and of choice signals, and finally the dynamical evolution of neural population activity and the distributions of (pairwise) neural interactions. This approach reveals shared patterns of statistical organization across cortex, including: (i) long-tailed distributions of activity, where quasi-silence seems to be the rule for a majority of neurons; that are barely distinguishable between spontaneous and active states; (ii) distributions of tuning parameters for sensory (and motor) variables, which show an extensive extrapolation and fragmentation of their representations in the periphery; and (iii) population-wide dynamics that reveal rotations of internal representations over time, whose traces can be found both in stimulus-driven and internally generated activity. We discuss how these insights are leading us away from the notion of discrete classes of cells, and are acting as powerful constraints on theories and models of cortical organization and population coding.
We study subexponential tail asymptotics for the distribution of the maximum Mt≔supu∈0,tXu of a process Xt with negative drift for the entire range of t>0. We consider compound renewal processes with ...linear drift and Lévy processes. For both processes we also formulate and prove the principle of a single big jump for their maxima. The class of compound renewal processes with drift particularly includes the Cramér–Lundberg renewal risk process.
A nonlinear random walk related to the porous medium equation (nonlinear Fokker–Planck equation) is investigated. This random walk is such that when the number of steps is sufficiently large, the ...probability of finding the walker in a certain position after taking a determined number of steps approximates to a q-Gaussian distribution ( G q , β ( x ) ∝ 1 − ( 1 − q ) β x 2 1 / ( 1 − q ) ), which is a solution of the porous medium equation. This can be seen as a verification of a generalized central limit theorem where the attractor is a q-Gaussian distribution, reducing to the Gaussian one when the linearity is recovered ( q → 1 ). In addition, motivated by this random walk, a nonlinear Markov chain is suggested.
We aim to derive a phenomenological approach to link the theories of anomalous transport governed by fractional calculus and stochastic theory with the conductivity behavior governed by the ...semi-empirical conductivity formalism involving Debye, Cole–Cole, Cole–Davidson, and Havriliak–Negami type conductivity equations. We want to determine the anomalous transport processes in the amorphous semiconductors and insulators by developing a theoretical approach over some mathematical instruments and methods. In this paper, we obtain an analytical expression for the average behavior of conductivity in complex or disordered media via the fractional-stochastic differential equation, the Fourier–Laplace transform, some natural boundary initial conditions, and familiar physical relations. We start with the stochastic equation of motion called the Langevin equation, develop its equivalent master equation called the Klein–Kramers or Fokker–Planck equation, and consider the time-fractional generalization of the master equation. Once we derive the fractional master equation, we then determine the expressions for the mean value of the variables or observables through some calculations and conditions. Finally, we use these expressions in the current density relation to obtain the average conductivity behavior.
Wildlife management often requires accurate estimates of hunter participation and species harvested. Sending pre-season hunting cards for recording activity can improve the accuracy of estimates. ...This research note examined the influence of record cards on the shape of waterfowl hunting survey response distributions. Data were obtained from a 2013-2014 mail survey of 1,796 waterfowl hunters in Illinois. Results indicated that individuals who received a record card were: (a) more likely to report small harvest and days hunting values (beginning of distribution), and (b) less likely to give responses that contributed to heaping (middle of distribution) compared to non-record card recipients. Record cards did not influence the end of the distribution, as frequency functions for all respondents were long-tailed distributions. Results imply that record card responses are more accurate than non-record card responses. Non-record card estimates were approximately 10% biased. Results supported continued use of record cards to limit bias.
Long-tailed distributions can distort findings, influence statistical tests, and result in small effect sizes. This research note proposed a definition of long-tailed distributions (i.e., SD/M ≥ 1) ...and developed an alternative formulation of the Cohen's d effect size based on percent differences. Three hypotheses were examined: (a) waterfowl hunter harvest distributions tend to be long-tailed distributions, (b) differences in the means of two long-tailed distributions have minimal (d < .2) effect sizes unless the percent difference exceeds 20%, and (c) a minimal effect size does not necessarily imply that the difference in means should be ignored. Data obtained from 29 (1990-2018) annual waterfowl surveys in Illinois (n = 45,978) supported all three hypotheses. Statistical and managerial implications are discussed.
In this paper, we investigate the precise large deviations for sums of φ-mixing and UND random variables with long-tailed distributions. The asymptotic relations for non random sum and random sum of ...random variables with long-tailed distributions are obtained.
PDF
