Semi-supervised multi-label classification using an extended graph-based manifold regularization

Graph-based algorithms are known to be effective approaches to semi-supervised learning. However, there has been relatively little work on extending these algorithms to the multi-label classification case. We derive an extension of the Manifold Regularization algorithm to multi-label classification, which is significantly simpler than the general Vector Manifold Regularization approach. We then augment our algorithm with a weighting strategy to allow differential influence on a model between instances having ground-truth vs. induced labels. Experiments on four benchmark multi-label data sets show that the resulting algorithm performs better overall compared to the existing semi-supervised multi-label classification algorithms at various levels of label sparsity. Comparisons with state-of-the-art supervised multi-label approaches (which of course are fully labeled) also show that our algorithm outperforms all of them even with a substantial number of unlabeled examples.

Method of an unmanned aerial vehicle composition route in space

The article is devoted to the development of a method of composition of the route of an unmanned aerial vehicle in three-dimensional space. The main difference of the presented method is the complex consideration of the features of the environment, which reflects the possible obstacles (active or passive) and other limitations of the problem when composing the route of the unmanned aerial vehicle in three-dimensional space. This allowed to increase the safety of the task in autonomous flight conditions. The article analyzes the main approaches to the composition of unmanned aerial vehicle routes in space. The conclusion about the shortcomings of the two-dimensional representation is made. The method presents four stages of the task. This is the stage of modeling the environment that reflects possible obstacles (active or passive) and other limitations of the task. Stage of construction of an extended graph of unmanned aerial vehicle routes in space. The difference of this stage is the adaptive consideration of the spatial location of active obstacles in space. The next stage is the route search stage, which connects the starting point with the end and bypasses all obstacles and allows you to build a starting route in the form of a broken line, which is formed by a sequence of waypoints, and connects the starting point with the end, bypassing obstacles. The last is the stage of obtaining the final result, which is provided by smoothing the obtained broken line. In this part of the composition method, to solve the problem of smoothing the trajectory of the unmanned aerial vehicle in space on the selected route, the expediency of using the method of non-uniform cubic B-spline is proved. With the help of this method the task of selection and optimization of the smoothing parameter is set and solved.

Extended Graph of the Fuzzy Topographic Topological Mapping Model

Fuzzy topological topographic mapping (FTTM) is a mathematical model which consists of a set of homeomorphic topological spaces designed to solve the neuro magnetic inverse problem. A sequence of FTTM, FTTMn, is an extension of FTTM that is arranged in a symmetrical form. The special characteristic of FTTM, namely the homeomorphisms between its components, allows the generation of new FTTM. The generated FTTMs can be represented as pseudo graphs. A graph of pseudo degree zero is a special type of graph where each of the FTTM components differs from the one adjacent to it. Previous researchers have investigated and conjectured the number of generated FTTM pseudo degree zero with respect to n number of components and k number of versions. In this paper, the conjecture is proven analytically for the first time using a newly developed grid-based method. Some definitions and properties of the novel grid-based method are introduced and developed along the way. The developed definitions and properties of the method are then assembled to prove the conjecture. The grid-based technique is simple yet offers some visualization features of the conjecture.

EGAT: Extended Graph Attention Network for Pedestrian Trajectory Prediction

To improve foresight and make correct judgment in advance, pedestrian trajectory prediction has a wide range of application values in autonomous driving, robot interaction, and safety monitoring. However, most of the existing methods only focus on the interaction of local pedestrians according to distance, ignoring the influence of far pedestrians; the range of network input (receptive field) is small. In this paper, an extended graph attention network (EGAT) is proposed to increase receptive field, which focuses not only on local pedestrians, but also on those who are far away, to further strengthen pedestrian interaction. In the temporal domain, TSG-LSTM (TS-LSTM and TG-LSTM) and P-LSTM are proposed based on LSTM to enhance information transmission by residual connection. Compared with state-of-the-art methods, the model EGAT achieves excellent performance on both ETH and UCY public datasets and generates more reliable trajectories.

The Consolidated Mathews Stability Graph for Open Stope Design

The stability graph method of stope design is one of the most widely used methods of stability assessments of stopes in underground polymetallic mines. The primary objective of this work is to introduce a new stability chart, which includes all relevant case histories, and to exclude parameters with uncertainties in the determination of stability number. The modified stability number was used to achieve this goal, and the Extended Mathews database was recalculated and compared with the new stability graph. In this study, a new refined Consolidated stability graph was developed by excluding the entry mining methods data from the Extended graph data, and only the non-entry methods data was used. The applicability of the proposed Consolidated stability chart was demonstrated by an open stope example. The stability for each stope surface was evaluated by a probabilistic approach employing a logistic regression model and the developed Consolidated stability chart. Comparing the stability analysis results with that of other published works of the same example shows that the determined Consolidated chart, in which the entry-method data is excluded, produces a more conservative and safer design. In conclusion, the size and quality of the dataset dictate the reliability of this approach.