Depth-first Research Articles

A Target Detection Algorithm for Remote Sensing Images Based on Deep Learning.

In order to improve the accuracy of remote sensing image target detection, this paper proposes a remote sensing image target detection algorithm DFS based on deep learning. Firstly, dimension clustering module, loss function, and sliding window segmentation detection are designed. The data set used in the experiment comes from GoogleEarth, and there are 6 types of objects: airplanes, boats, warehouses, large ships, bridges, and ports. Training set, verification set, and test set contain 73490 images, 22722 images, and 2138 images, respectively. It is assumed that the number of detected positive samples and negative samples is A and B, respectively, and the number of undetected positive samples and negative samples is C and D, respectively. The experimental results show that the precision-recall curve of DFS for six types of targets shows that DFS has the best detection effect for bridges and the worst detection effect for boats. The main reason is that the size of the bridge is relatively large, and it is clearly distinguished from the background in the image, so the detection difficulty is low. However, the target of the boat is very small, and it is easy to be mixed with the background, so it is difficult to detect. The MAP of DFS is improved by 12.82%, the detection accuracy is improved by 13%, and the recall rate is slightly decreased by 1% compared with YOLOv2. According to the number of detection targets, the number of false positives (FPs) of DFS is much less than that of YOLOv2. The false positive rate is greatly reduced. In addition, the average IOU of DFS is 11.84% higher than that of YOLOv2. For small target detection efficiency and large remote sensing image detection, the DFS algorithm has obvious advantages.

BDPS: An Efficient Spark-Based Big Data Processing Scheme for Cloud Fog-IoT Orchestration

The Internet of Things (IoT) has seen a surge in mobile devices with the market and technical expansion. IoT networks provide end-to-end connectivity while keeping minimal latency. To reduce delays, efficient data delivery schemes are required for dispersed fog-IoT network orchestrations. We use a Spark-based big data processing scheme (BDPS) to accelerate the distributed database (RDD) delay efficient technique in the fogs for a decentralized heterogeneous network architecture to reinforce suitable data allocations via IoTs. We propose BDPS based on Spark-RDD in fog-IoT overlay architecture to address the performance issues across the network orchestration. We evaluate data processing delays from fog-IoT integrated parts using a depth-first-search-based shortest path node finding configuration, which outperforms the existing shortest path algorithms in terms of algorithmic (i.e., depth-first search) efficiency, including the Bellman–Ford (BF) algorithm, Floyd–Warshall (FW) algorithm, Dijkstra algorithm (DA), and Apache Hadoop (AH) algorithm. The BDPS exhibits low latency in packet deliveries as well as low network overhead uplink activity through a map-reduced resilient data distribution mechanism, better than in BF, DA, FW, and AH. The overall BDPS scheme supports efficient data delivery across the fog-IoT orchestration, outperforming faster node execution while proving effective results, compared to DA, BF, FW and AH, respectively.

Improving RGB-D SLAM in dynamic environments using semantic aided segmentation

SummaryMost conventional simultaneous localization and mapping (SLAM) approaches assume the working environment to be static. In a highly dynamic environment, this assumption divulges the impediments of a SLAM algorithm that lack modules that distinctively attend to dynamic objects despite the inclusion of optimization techniques. This work exploits such environments and reduces the effects of dynamic objects in a SLAM algorithm by separating features belonging to dynamic objects and static background using a generated binary mask image. While the features belonging to the static region are used for performing SLAM, the features belonging to non-static segments are reused instead of being eliminated. The approach employs deep neural network or DNN-based object detection module to obtain bounding boxes and then generates a lower resolution binary mask image using depth-first search algorithm over the detected semantics, characterizing the segmentation of the foreground from the static background. In addition, the features belonging to dynamic objects are tracked into consecutive frames to obtain better masking consistency. The proposed approach is tested on both publicly available dataset as well as self-collected dataset, which includes both indoor and outdoor environments. The experimental results show that the removal of features belonging to dynamic objects for a SLAM algorithm can significantly improve the overall output in a dynamic scene.

Exploring Graph Traversal Algorithms in Graph-Based Molecular Generation.

Here, we explore the impact of different graph traversal algorithms on molecular graph generation. We do this by training a graph-based deep molecular generative model to build structures using a node order determined via either a breadth- or depth-first search algorithm. What we observe is that using a breadth-first traversal leads to better coverage of training data features compared to a depth-first traversal. We have quantified these differences using a variety of metrics on a data set of natural products. These metrics include percent validity, molecular coverage, and molecular shape. We also observe that by using either a breadth- or depth-first traversal it is possible to overtrain the generative models, at which point the results with either graph traversal algorithm are identical.

Efficient approximate approach for graph edit distance problem

• Near optimal results, with a less than 5% deviation from optimal results of a DFS algorithm. • Fast execution time. • Scalable since it can deal with large data-sets in small execution time. • report tight upper-bounds for large data-sets. Graph Edit Distance (GED) problem is a well-known tool used to measure the similarity/dissimilarity between two graphs. It searches for the best set of edit operations (in terms of cost) that transforms one graph into another. Due to the NP-hardness nature of the problem, the search space increases exponentially making exact approaches impossible to use for large graphs. In this context, there is a huge need for approaches that give near-optimal results in reasonable time. In this paper, we propose a tree-based approximate approach for dealing with GED problem. It operates on a search tree that models all possible solutions of the problem. Since exploring the whole tree is impractical; this approach keeps only the best k nodes at each level of the tree for further exploration. This reduces enormously the execution time without scarifying the solution quality. Experiments using small and medium size data-sets show the low deviation of our results as compared to the optimal results of a Depth First Search algorithm. Moreover, our approach show a strong scalability potential by dealing with large data-sets in low execution time.

Reliable Route Selection for Wireless Sensor Networks with Connection Failure Uncertainties.

For wireless sensor networks (WSN) with connection failure uncertainties, traditional minimum spanning trees are no longer a feasible option for selecting routes. Reliability should come first before cost since no one wants a network that cannot work most of the time. First, reliable route selection for WSNs with connection failure uncertainties is formulated by considering the top-k most reliable spanning trees (RST) from graphs with structural uncertainties. The reliable spanning trees are defined as a set of spanning trees with top reliabilities and limited tree weights based on the possible world model. Second, two tree-filtering algorithms are proposed: the k minimum spanning tree (KMST) based tree-filtering algorithm and the depth-first search (DFS) based tree-filtering algorithm. Tree-filtering strategy filters the candidate RSTs generated by tree enumeration with explicit weight thresholds and implicit reliability thresholds. Third, an innovative edge-filtering method is presented in which edge combinations that act as upper bounds for RST reliabilities are utilized to filter the RST candidates and to prune search spaces. Optimization strategies are also proposed for improving pruning capabilities further and for enhancing computations. Extensive experiments are conducted to show the effectiveness and efficiency of the proposed algorithms.

Approaching the bi-objective critical node detection problem with a smart initialization-based evolutionary algorithm.

Determining the critical nodes in a complex network is an essential computation problem. Several variants of this problem have emerged due to its wide applicability in network analysis. In this article we study the bi-objective critical node detection problem (BOCNDP), which is a new variant of the well-known critical node detection problem, optimizing two objectives at the same time: maximizing the number of connected components and minimizing the variance of their cardinalities. Evolutionary multi-objective algorithms (EMOA) are a straightforward choice to solve this type of problem. We propose three different smart initialization strategies which can be incorporated into any EMOA. These initialization strategies take into account the basic properties of the networks. They are based on the highest degree, random walk (RW) and depth-first search. Numerical experiments were conducted on synthetic and real-world network data. The three different initialization types significantly improve the performance of the EMOA.

Autosetting soliton pulsation in a fiber laser by an improved depth-first search algorithm.

Soliton pulsation is one of the most fascinating phenomena in ultrafast fiber lasers, owing to its rich nonlinear dynamics and potential generation of high peak power pulse. However, it is still a challenge to efficiently search for pulsating soliton in fiber lasers because it requires a fine setting of laser cavity parameters. Here, we report the autosetting soliton pulsation in a passively mode-locked fiber laser. The parameters of electronic polarization controller are intelligently adjusted to search for pulsating soliton state by the improved depth-first search algorithm. Moreover, the intensity modulation depth of pulsating soliton could be flexibly controlled. These findings indicate that the intelligent control of a fiber laser is an effective way to explore on-demand soliton dynamics and is also beneficial to the optimization of ultrafast laser performance.

Parallel Computing for Multi-Objective Train Rescheduling

In railway traffic systems, it is essential to achieve a high punctuality to satisfy the goals of the involved stakeholders. Thus, whenever disturbances occur, it is important to effectively reschedule trains while considering the perspectives of various stakeholders. This typically involves solving a multi-objective train rescheduling problem, which is much more complex than its single-objective counterpart. Solving such a problem in real time for practically relevant problem sizes is computationally challenging. The reason is that the rescheduling solution(s) of interest are dispersed across a large search tree. The tree needs to be navigated fast while pruning off branches leading to undesirable solutions and exploring branches leading to potentially desirable solutions. The use of parallel computing enables such a fast navigation of the tree. This article presents a heuristic parallel algorithm to solve the multi-objective train rescheduling problem. The parallel algorithm combines a depth-first search with simultaneous breadth-wise tree exploration while searching the tree for solutions. An existing parallel algorithm for single-objective train rescheduling has been redesigned, primarily, by (i) pruning based on multiple metrics, and (ii) maintaining a set of upper bounds. The redesign improved the quality of the obtained rescheduling solutions and showed better speedups for several disturbance scenarios.

Information Set Generation in Partially Observable Games

We address the problem of making single-point decisions in large partially observable games, where players interleave observation, deliberation, and action. We present information set generation as a key operation needed to reason about games in this way. We show how this operation can be used to implement an existing decision-making algorithm. We develop a constraint satisfaction algorithm for performing information set generation and show that it scales better than the existing depth-first search approach on multiple non-trivial games.

Max Is More than Min: Solving Maximization Problems with Heuristic Search

Most work in heuristic search considers problems where a low cost solution is preferred (MIN problems). In this paper, we investigate the complementary setting where a solution of high reward is preferred (MAX problems). Example MAX problems include finding the longest simple path in a graph, maximal coverage, and various constraint optimization problems. We examine several popular search algorithms for MIN problems — optimal, suboptimal, and bounded suboptimal - and discover the curious ways in which they misbehave on MAX problems. We propose modifications that preserve the original intentions behind the algorithms but allow them to solve MAX problems, and compare them theoretically and empirically. Interesting results include the failure of bidirectional search and a discovered close relationships between Dijkstra's algorithm, weighted A*, and depth-first search. This work demonstrates that MAX problems demand their own heuristic search algorithms, which are worthy objects of study in their own right.

SANE: A sequence combined attentive network embedding model for COVID-19 drug repositioning

The COVID-19 has now spread all over the world and causes a huge burden for public health and world economy. Drug repositioning has become a promising treatment strategy in COVID-19 crisis because it can shorten drug development process, reduce pharmaceutical costs and reposition approval drugs. Existing computational methods only focus on single information, such as drug and virus similarity or drug–virus network feature, which is not sufficient to predict potential drugs. In this paper, a sequence combined attentive network embedding model SANE is proposed for identifying drugs based on sequence features and network features. On the one hand, drug SMILES and virus sequence features are extracted by encoder–decoder in SANE as node initial embedding in drug–virus network. On the other hand, SANE obtains fields for each node by attention-based Depth-First-Search (DFS) to reduce noises and improve efficiency in representation learning and adopts a bottom-up aggregation strategy to learn node network representation from selected fields. Finally, a forward neural network is used for classifying. Experiment results show that SANE has achieved the performance with 81.98% accuracy and 0.8961 AUC value and outperformed state-of-the-art baselines. Further case study on COVID-19 indicates that SANE has a strong predictive ability since 25 of the top 40 (62.5%) drugs are verified by valuable dataset and literatures. Therefore, SANE is powerful to reposition drugs for COVID-19 and provides a new perspective for drug repositioning.

Automatic Move Pruning Revisited

In this paper we show that the move pruning method we presented at SoCS last year sometimes prunes all the least-cost paths from one state to another. We present two examples exhibiting this erroneous behaviour--a simple, artificial example and a slightly more complex example that arose in last year's experiments. We then formally prove that a simple modification to our move pruning method makes it "safe," i.e., it will never prune all the least-cost paths between a pair of states. Finally, we present the results of rerunning last year's experiments with this provably safe version of move pruning and show that last year's main conclusions still hold, namely: (1) in domains where there are short redundant sequences move pruning produces substantial speedups in depth-first search, and (2) in domains where the only short redundant sequences are 2-cycles, move pruning is faster than parent pruning by a factor of two or more.

Diverse Depth-First Search in Satisificing Planning

Diverse best-first search (DBFS) is a successful algorithm for satisficing planning that is built on top of greedy best-first search and uses stochastic search techniques to select the next ``best'' state that is not always in agreement with the heuristic function. DBFS thus considers a diversity of search directions and avoids search plateaus caused by inaccurate heuristic estimation. However, the scalability of DBFS is still limited by the available memory resource, as in any planning system that uses best-first search at its core. This paper presents the diverse depth-first search (DDFS) algorithm that overcomes the memory bottleneck of DBFS. DDFS requires much less memory than DBFS yet it successfully incorporates DBFS' randomized node selection scheme for effective search diversification. Additionally, DDFS is enhanced with a transposition table to avoid duplicate search effort commonly found in depth-first search. Experimental results in solving problems used in previous International Planning Competitions show that DDFS can solve instances that remain unsolvable by DBFS due to its excessive memory requirements.

An Optimal Algorithm for Planning Shearer Trailing Drum Cutting Path

The intelligent adaptive cutting of the shearer is one of the key technologies to realize the intelligent working face. However, since the “memory cut” technology is the mainstream technology, which cannot actively adapt to the coal seam variations, the trailing drum usually cuts at a fixed height without a planned cutting path. This paper analyzes the shearer’s location characteristics before and after the advancement to propose a complete calculation method for the advancing path of the shearer, which simulates all of its possible advancing paths for subsequent n cuttings. The multitree and depth-first search algorithms are utilized to filter out the optimal advance path under different mining requirements. Simultaneously, this paper indicates that the vertical curvature of the armored face conveyor (AFC) should be considered in the calculation process of the optimal advancing path at different positions of the working face to obtain the shearer’s planned cutting path for subsequent n cuttings. The proposed algorithm in this paper has apparent advantages over the “memory cut” technology and provides a good solution for the intelligent planning of cutting and pitch steering of the shearers.

Best-First Search for Bounded-Depth Trees

Tree search is a common technique for solving constraint satisfaction and combinatorial optimization problems. The most popular strategies are depth-first search and limited discrepancy search. Aside from pruning or ordering the children of each node, these algorithms do not adapt their search order to take advantage of information that becomes available during search, such as heuristic scores or leaf costs. We present a framework called best-leaf-first search (BLFS) that uses this additional information to estimate the cost of taking discrepancies in the search tree and then attempts to visit leaves in a best-first order. In this way, BLFS brings the idea of best-first search from shortest path problems to the areas of constraint satisfaction and combinatorial optimization. Empirical results demonstrate that this new dynamic approach results in better search performance than previous static search strategies on two very different domains: structured CSPs and the traveling salesman problem.

Anytime AND/OR Depth-First Search for Combinatorial Optimization

One popular and efficient scheme for solving exactly combinatorial optimization problems over graphical models is depth-first Branch and Bound. However, when the algorithm exploits problem decomposition using AND/OR search spaces, its anytime behavior breaks down. This paper 1) analyzes and demonstrates this inherent conflict between effective exploitation of problem decomposition (through AND/OR search spaces) and the anytime behavior of depth-first search (DFS), 2) presents a first scheme to address this issue while maintaining desirable DFS memory properties, 3) analyzes and demonstrates its effectiveness. Our work is applicable to any problem that can be cast as search over an AND/OR search space.

HANP-Miner: High average utility nonoverlapping sequential pattern mining

Nonoverlapping sequential pattern mining (SPM) is a data analysis task, which aims at identifying repetitive sequential patterns with gap constraint in a set of discrete sequences. Nonoverlapping means that any character in the sequence can be rematched by characters at different positions in the pattern, while overlapping means that any character can be reused at the same position, which can be regarded as no condition. Compared with overlapping SPM, nonoverlapping SPM has more strict constraint on the occurrences of pattern, and meets the Apriori property. However, current algorithms mine frequent or closed patterns, resulting in some low-frequency but extremely important patterns being ignored. To tackle this issue, this paper proposes to mine high average utility nonoverlapping sequential patterns (HANP). An efficient algorithm called HANP-Miner is proposed, which involves two key steps: support calculation and candidate pattern reduction. To calculate the support (the occurrence frequency of a pattern), depth-first search and backtracking strategies based on the simplified Nettree structure are adopted, an approach that effectively reduces the time and space complexities of the algorithm. To effectively reduce the number of candidate patterns, a pattern join strategy based on an upper bound on the average utility is proposed. Experiments on biological sequences and real sales sequences are carried out to verify the efficiency of HANP-Miner and the superiority of HANPs. The results demonstrate that HANP-Miner is not only more efficient, but that the HANPs mined in this way are also more valuable than existing frequent patterns. The algorithms and datasets can be downloaded from https://github.com/wuc567/Pattern-Mining/tree/master/HANP-Miner.

End vertices of graph searches on bipartite graphs

For a graph search algorithm, the end vertex problem is concerned with which vertices of a graph can be the last visited by this algorithm. We show that for both lexicographic depth-first search and maximum cardinality search, the end vertex problem is NP-complete on bipartite graphs, even if the maximum degree of the graph is bounded.

Behavior anomaly detection based on big data analysis of Internet of Things

The technical requirements of behavior anomaly detection are higher and higher. Using the Internet of things technology combined with a variety of big data analysis algorithms, we can achieve accurate behavior anomaly detection by classifying behavior data sets to a large extent. In this paper, PLA - PRF (parallel random forest) algorithm is used to realize the behavior anomaly detection model of Internet of things integrating big data analysis. In behavior detection, the PRF algorithm and DFS algorithm are compared in the case of a different number of decision trees. The results show that, compared with DRF algorithm, PLA-PRF, SPARK MLRF(Spark Machine Learning Random Forests) and PRF algorithm perform better on the four datasets, with kappa values increased by about 3.13%, 2.56% and 1.98% respectively. In contrast, PLA-PRF algorithm has higher accuracy in the case of a small sample size. With the increase of sample size, the accuracy of behavior anomaly detection gradually decreases; because the algorithm is in subspace in the process of construction, some high pheromone features are abandoned, which makes the new spatial information of features insufficient, resulting in the decision tree training process does not learn the inherent laws of abandoned data. Compared with spark MLRF and DRF, PLA-PRF has a faster execution speed in large data sets, and with the increase of data volume, the advantage is more prominent. This is because PLA-PRF uses data reuse strategy "DRS" in the process of parallelization, which reduces the data communication overhead in a distributed environment and improves the parallelization efficiency of the algorithm.