基于DSPP的单目图像深度估计

空洞空间金字塔池化（ASPP）在深度学习各种任务中均有应用,传统ASPP模块只考虑了提升卷积感受视野,但ASPP中的每次空洞卷积选取的像素点分散,会丢失大量像素点间的信息,而深度估计属于密集预测任务。针对ASPP模块这一弊端提出了一种动态密集的DSPP模块。该模块用一种动态卷积代替空洞卷积,结合ASPP的思想,采用不同大小的卷积尺寸,并结合通道注意力充分利用每一层的特征,解决了ASPP丢失信息的问题,与ASPP相比在大大减小模块参数量的前提下,提升了整体模型的准确率。在NYU Depth v2数据集上与主流算法相比深度图在均方根误差（RMSE）上降低了12.5%,到0.407,并且准确率（δ<1.25）提高了3.4%,达到0.875,验证了算法的有效性。     

基于极限学习机的堤防工程多元风险指标评价方法

为提高堤防工程风险评价的准确性,提出了基于极限学习机的堤防工程多元风险指标评价方法。首先,综合考虑影响堤防风险的28个评价指标,利用层次分析法从预警系统、堤防工程系统、环境系统和社会经济系统这4个方面建立了堤防工程多元风险评价指标体系。接着,基于极限学习机算法对28个指标进行标准化处理及分级标准构建,以风险指标作为输入量,分级隶属度作为输出量,划分风险等级,量化评价指标,估计多元风险评价指标值和判断风险的严重程度。最后,依托鄱阳湖重点堤防——康山大堤,构建多元风险评价指标体系,运用极限学习机算法计算多元风险评价指标值。评价结果表明:康山大堤目前处于基本安全水平,符合康山大堤经过两次加固后的工程实际情况,并与其他方法进行对比,验证了提出方法的可靠性和有效性。该方法可拓展应用到其他重要水工结构工程风险评估中。     

Probabilistic heuristic estimates

Though they constitute the major knowledge source in problem-solving systems, no unified theory of heuristics has emerged. Pearl [15] defines heuristics as criteria, methods, or principles for deciding which among several alternative courses of action promises to be the most effective in order to achieve some goal. The absence of a more precise definition has impeded our efforts to understand, utilize, and discover heuristics. Another consequence is that problem-solving techniques which rely on heuristic knowledge cannot be relied upon to act rationally — in the sense of the normative theory of rationality.To provide a sound basis for BPS, the Bayesian Problem-Solver, we have developed a simple formal theory of heuristics, which is general enough to subsume traditional heuristic functions as well as other forms of problem-solving knowledge, and to straddle disparate problem domains. Probabilistic heuristic estimates represent a probabilistic association of sensations with prior experience — specifically, a mapping from observations directly to subjective probabilities which enables the use of theoretically principled mechanisms for coherent inference and decision making during problem-solving. This paper discusses some of the implications of this theory, and describes its successful application in BPS.This research was made possible by support from Heuristicrats, the National Aeronautics and Space Administration, and the Rand Corporation.     

LPNet: A DNN based latency prediction technique for application mapping in Network-on-Chip design

Analytical models used for latency estimation of Network-on-Chip (NoC) are not producing reliable accuracy. This makes these analytical models difficult to use in optimization of design space exploration. In this paper, we propose a learning based model using deep neural network (DNN) for latency predictions. Input features for DNN model are collected from analytical model as well as from Booksim simulator. Then this DNN model has been adopted in mapping optimization loop for predicting the best mapping of given application and NoC parameters combination. Our simulations show that using the proposed DNN model, prediction error is less than 12% for both synthetic and application specific traffic. More than 108 times speedup could be achieved using DPSO with DNN model compared to DPSO using Booksim simulator.     

Environmental Perception with Spatial Regularization Correlation Filter for Visual Tracking

With the introduction of correlation filtering (CF), the performance of visual object tracking is significantly improved. Circular shifts collecting samples is a key component of the CF tracker, and it also causes negative boundary effects. Most trackers add spatial regularization to alleviate boundary effects well. However, these trackers ignore the effect of environmental changes on tracking performance, and the filter discriminates poorly in the background interference. Here, to break these limitations, we propose a new correlation filter model, namely Environmental Perception with Spatial Regularization Correlation Filter for Visual Tracking. Specifically, we use the Average Peak to Correlation Energy (APCE) and the response value error between the two frames together to perceive environmental changes, which adjusts the learning rate to make the template more adaptable to environmental changes. To enhance the discriminatory capability of the filter, we use real background information as negative samples to train the filter model. In addition, the introduction of the regular term destroys the closed solution of CF, and this problem can be effectively solved by the use of the alternating direction method of multipliers (ADMM). Extensive experimental evaluations on three large tracking benchmarks are performed, which demonstrate the good performance of the proposed method over some of the state-of-the-art trackers.     

No-reference stereoscopic image quality assessment using quaternion wavelet transform and heterogeneous ensemble learning

As the demand for high-quality stereo images has grown in recent years, stereoscopic image quality assessment (SIQA) has become an important research area in modern image processing technology.In this paper, we propose a no-reference stereoscopic image quality assessment (NR-SIQA) model using heterogeneous ensemble learning ‘quality-aware’ features from luminance image, chrominance image, disparity and cyclopean images via quaternion wavelet transform (QWT). Firstly, luminance image and chrominance image are generated by CIELAB color space as monocular perception, and the novel disparity and cyclopean images are utilized to complement with monocular information. Then, a number of ‘quality-aware’ features in the quaternion wavelet domain are discovered, including entropy, texture features, energy features, energy differences features and MSCN coefficients of high frequency sub-band. Finally, a heterogeneous ensemble model via support vector regression (SVR) & extreme learning machine (ELM) & random forest (RF) is proposed to predict quality score, and bootstrap sampling and rotated feature space are used to increase the diversity of data distribution. Comparing with the state-of-the-art NR-SIQA models, experimental results on four public databases prove the accuracy and robustness of the proposed model.     

Quadratic polynomial guided fuzzy C-means and dual attention mechanism for medical image segmentation

Medical image segmentation is the most complex and important task in the field of medical image processing and analysis, as it is linked to disease diagnosis accuracy. However, due to the medical image's high complexity and noise, segmentation performance is limited. We propose a novel quadratic polynomial guided fuzzy C-means and dual attention mechanism composite network model architecture to address the aforementioned issues (QPFC-DA). It has mechanisms for channel and spatial edge attention, which guide the content and edge segmentation branches, respectively. The bi-directional long short-term memory network was added after the two content segmentation branches to better integrate multi-scale features and prevent the loss of important features. Furthermore, the fuzzy C-means algorithm guided by the quadratic polynomial can better distinguish the image's weak edge regions and has a degree of noise resistance, resulting in a membership matrix with less ambiguity and a more reliable segmentation result. We also conducted comparison and ablation experiments on three medical data sets. The experimental results show that this method is superior to several other well-known methods.     

Mine,ours, and yours: Whose engagement and prior knowledge affects individual achievement from online collaborative learning?

In computer-supported collaborative learning research, studies examining the combined effects of individual level, group level and within-group differences level measures on individual achievement are scarce. The current study addressed this by examining whether individual, group and within-group differences regarding engagement and prior knowledge predict individual achievement. Engagement was operationalised as group members' exhibited activities in the task space (i.e., discussing domain-content) and social space (i.e., regulating ideas, actions and socioemotional processes). Prior knowledge and achievement were operationalised as group members' performance on a domain-related pre-test and post-test, respectively. Data was collected for 95 triads of secondary education students collaborating on a complex business-economics problem. Subsequently, three different multilevel models were tested to examine the combined effect. First a model with the individual level measures (model 1) was tested and in subsequent models the group level measures (model 2) and within-group levels measures (model 3) were added. Findings indicate model 2 showed the best fit; group members' individual engagement in the social space activities as well as the groups' average prior knowledge positively predicts individual achievement. No effects were found for either group members' or groups' engagement in the task space and for the within-group differences.     

Maximizing learning without sacrificing the fun: Stealth assessment,adaptivity and learning supports in educational games

In this study, we investigated the validity of a stealth assessment of physics understanding in an educational game, as well as the effectiveness of different game-level delivery methods and various in-game supports on learning. Using a game called Physics Playground, we randomly assigned 263 ninth- to eleventh-grade students into four groups: adaptive, linear, free choice and no-treatment control. Each condition had access to the same in-game learning supports during gameplay. Results showed that: (a) the stealth assessment estimates of physics understanding were valid—significantly correlating with the external physics test scores; (b) there was no significant effect of game-level delivery method on students' learning; and (c) physics animations were the most effective (among eight supports tested) in predicting both learning outcome and in-game performance (e.g. number of game levels solved). We included student enjoyment, gender and ethnicity in our analyses as moderators to further investigate the research questions.     

Inquiry-based mobile learning in secondary school science education: A systematic review

Recent years have seen a growing call for inquiry-based learning in science education, and mobile technologies are perceived as increasingly valuable tools to support this approach. However, there is a lack of understanding of mobile technology-supported inquiry-based learning (mIBL) in secondary science education. More evidence-based, nuanced insights are needed into how using mobile technologies might facilitate students' engagement with various levels of inquiry and enhance their science learning. We, therefore, conducted a robust systematic literature review (SLR) of the research articles on mIBL in secondary school science education that have been published from 2000 to 2019. We reviewed and analysed 31 empirical studies (34 articles) to explore the types of mIBL, and the benefits and constraints of mIBL in secondary school science education. The findings of this SLR suggest new research areas for further exploration and provide implications for science teachers' selection, use and design of mIBL approaches in their teaching.