Search

The base learners and labeled samples (shots) in an ensemble few-shot classifier greatly affect the model performance. When the performance is not satisfactory, it is usually difficult to understand the underlying causes and make improvements. To tackle this issue, we propose a visual analysis method, FSLDiagnotor. Given a set of base learners and a collection of samples with a few shots, we consider two problems: 1) finding a subset of base learners that well predict the sample collections; and 2) replacing the low-quality shots with more representative ones to adequately represent the sample collections. We formulate both problems as sparse subset selection and develop two selection algorithms to recommend appropriate learners and shots, respectively. A matrix visualization and a scatterplot are combined to explain the recommended learners and shots in context and facilitate users in adjusting them. Based on the adjustment, the algorithm updates the recommendation results for another round of improvement. Two case studies are conducted to demonstrate that FSLDiagnotor helps build a few-shot classifier efficiently and increases the accuracy by 12% and 21%, respectively. [ABSTRACT FROM AUTHOR]

Relational topic models (RTMs) provide a probabilistic generative process to describe both the link structure and document contents for document networks, and they have shown promise on predicting network structures and discovering latent topic representations. However, existing RTMs have limitations in both the restricted model expressiveness and incapability of dealing with imbalanced network data. To expand the scope and improve the inference accuracy of RTMs, this paper presents three extensions: 1) unlike the common link likelihood with a diagonal weight matrix that allows the-same-topic interactions only, we generalize it to use a full weight matrix that captures all pairwise topic interactions and is applicable to asymmetric networks; 2) instead of doing standard Bayesian inference, we perform regularized Bayesian inference (RegBayes) with a regularization parameter to deal with the imbalanced link structure issue in real networks and improve the discriminative ability of learned latent representations; and 3) instead of doing variational approximation with strict mean-field assumptions, we present collapsed Gibbs sampling algorithms for the generalized relational topic models by exploring data augmentation without making restricting assumptions. Under the generic RegBayes framework, we carefully investigate two popular discriminative loss functions, namely, the logistic log-loss and the max-margin hinge loss. Experimental results on several real network datasets demonstrate the significance of these extensions on improving prediction performance. [ABSTRACT FROM AUTHOR]

Because of the barrier lake affected by highly changing and complex environment, it is always a great challenge that proper actions must be performed within a limited amount of time. Thus, how to scientifically and efficiently analyze the dam-break risk of the barrier lake and its impact area is very important for barrier lake disposal and downstream communities transfer. To improve the efficiency and the accuracy of the dam-break flood routing simulation, this paper mainly focuses on the construction of the virtual geographic environment (VGE) system and the scale effect analysis. Unlike most of the current cellular automata (CA)-based flood simulation systems, the proposed VGE system can offer an intuitive, efficient, and interactive visualization environment through which users can explore complicated spatial information and conduct risk assessment work. Some key technologies including the CA model of dam-break flood, the VGE system framework, the impact analysis method, and the scale effect analysis method were discussed in detail. A prototype system was developed to support dam-break simulation and risk analysis of the Xiaojiaoqiao barrier lake in Anxian County, Sichuan Province of China. By means of a variety of cell scales effect analysis experiments, the adaptation scope and characteristics of multiscale cells were obtained to implement the simulation analysis of the dam-break flood routing better. The proposed VGE system can improve the efficiency of risk assessment and decision-making. [ABSTRACT FROM PUBLISHER]