Article
Computer Science, Artificial Intelligence
Tomas Kulvicius, Minija Tamosiunaite, Florentin Worgotter
Summary: This paper presents a method for solving path-finding problems by transforming cost values into synaptic weights in a neural network. The method allows for online weight adaptation using network learning mechanisms, and has been demonstrated to be effective in navigating in environments with obstacles and following specific sequences of path nodes.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Bojie Shen, Muhammad Aamir Cheema, Daniel D. Harabor, Peter J. Stuckey
Summary: The study explores optimal and suboptimal algorithms for the Euclidean Shortest Path Problem in two dimensions. Results show that the new algorithm outperforms other ESPP planners in terms of speed, performance, path quality, and flexibility.
ARTIFICIAL INTELLIGENCE
(2022)
Article
Mathematics
Fabrizio Magrini, Malcolm Sambridge
Summary: This article presents a theoretical framework that connects Fermat's principle of least time with optimal transport theory using a cost function that enforces local transport. The proposed method can be used to find shortest paths in media through optimal transport plans, and it provides physically significant solutions in both directed and undirected graphs. The approach offers computational advantages over traditional algorithms in terms of efficiency. The article also highlights potential research directions for further improving computational efficiency.
Article
Education & Educational Research
Marcelo Campo, Analia Amandi, Julio Cesar Biset
Summary: Moodle is a widely used platform for Virtual Learning Management Systems in educational institutions, providing a standard solution for teaching and learning. Researchers can conduct experimental research on learning styles through Moodle, but limitations in its design make it challenging to support dynamic computational behaviors such as teachbot implementations.
EDUCATION AND INFORMATION TECHNOLOGIES
(2021)
Article
Computer Science, Information Systems
Sofiane Abbar, Rade Stanojevic, Mashaal Musleh, Mohamed ElShrif, Mohamed Mokbel
Summary: QARTA is an open-source, highly accurate and scalable map service system that utilizes machine learning techniques to construct its own map and calibrate query answers based on contextual information. The demo showcases the efficiency and accuracy of QARTA in real-world applications.
PROCEEDINGS OF THE VLDB ENDOWMENT
(2021)
Article
Automation & Control Systems
Chase St. Laurent, Raghvendra Cowlagi
Summary: A coupled path-planning and sensor configuration method is proposed to minimize exposure to an unknown threat field. Gaussian Process regression is used to estimate the threat field from sensor measurements. The method introduces a task-driven information gain metric for sensor configuration and a surrogate metric for computational efficiency. It outperforms traditional decoupled information-driven sensor configuration in finding near-optimal plans.
Article
Automation & Control Systems
Vivek S. Borkar, Alexandre Reiffers-Masson
Summary: In this article, a variant of the classical DeGroot model of opinion propagation with random interactions is considered. The study focuses on a situation where a certain subset of agents can be controlled by a control parameter. The problem is mapped to a shortest path problem and analyzed using a nonclassical policy gradient scheme. The article also discusses the case when only certain interactions between agents are observed.
IEEE TRANSACTIONS ON CONTROL OF NETWORK SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Hamid Beigy, Mohammad Reza Meybodi
Summary: This paper presents an iterative stochastic algorithm for solving the stochastic shortest path problem, using distributed learning automata to dynamically determine which edges to sample. As the number of samples taken from the edges increases, the probability of finding the shortest path also increases. Experimental results validate the theoretical results on various stochastic and random graphs.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Information Systems
Omer Amar, Ilana Sarfati, Kobi Cohen
Summary: This study investigates the problem of adaptive routing in wireless communication networks using online learning. It aims to develop an algorithm to learn optimal paths for data transmission, maximizing network throughput without complete knowledge of link states. The proposed algorithm, OLSB, is designed based on a novel learning strategy and achieves efficient adaptive path selection. Theoretical analysis and extensive numerical simulations demonstrate its high efficiency and performance.
Article
Computer Science, Information Systems
Jing Ren, Xishi Huang, Raymond N. Huang
Summary: This paper proposes a novel deep reinforcement learning method for optimal path planning for mobile robots. By combining dynamic programming with deep reinforcement learning, the method overcomes the issues of slow learning process and poor training data quality, and achieves promising experimental results.
Article
Engineering, Civil
Zhongqiang Ren, Zachary B. Rubinstein, Stephen F. Smith, Sivakumar Rathinam, Howie Choset
Summary: The Resource Constrained Shortest Path Problem (RCSPP) aims to find a minimum-cost path between a start and a goal location while keeping the resource consumption within limits. Solving RCSPP is challenging due to the need to compare and maintain partial paths based on multiple criteria and the absence of a single path that optimizes all criteria simultaneously. This paper presents ERCA*, a fast algorithm based on A* that efficiently handles multiple resource constraints and outperforms existing algorithms in terms of runtime efficiency.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Computer Science, Hardware & Architecture
Enes Erdin, Mumin Cebe, Kemal Akkaya, Eyuphan Bulut, Selcuk Uluagac
Summary: This paper discusses the challenges of using Bitcoin in micropayments and proposes a solution by creating a private payment channel network to improve payment robustness and privacy. Using a heuristic approach and the k-shortest path algorithm, the paper provides a solution that comes close to an optimal outcome.
JOURNAL OF NETWORK AND COMPUTER APPLICATIONS
(2021)
Article
Automation & Control Systems
Cristian Consonni, Martin Brugnara, Paolo Bevilacqua, Anna Tagliaferri, Marco Frego
Summary: This paper explores the applicability of machine learning models and techniques to the Markov-Dubins path planning problem. The study finds that machine learning techniques have been applied to various fields, and their pervasiveness has led to significant improvements and support for specialized use cases. Although established numerical and optimization techniques already exist for the Markov-Dubins problem, there is a need to investigate the potential benefits of machine learning approaches in terms of speed-ups and application domains. The results show that machine learning approaches are comparable to state-of-the-art solutions, opening new avenues for interdisciplinary applications in planning problems.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Information Systems
Mengxuan Zhang, Lei Li, Xiaofang Zhou
Summary: Shortest path computation is essential for network applications, and the Dynamic Shortest Path (DSP) problem has garnered significant attention. Researchers identified dimensions of the DSP problem and evaluated state-of-the-art methods systematically in dynamic networks. The results can guide choosing the best solution during system implementation.
PROCEEDINGS OF THE VLDB ENDOWMENT
(2021)
Article
Computer Science, Artificial Intelligence
Lihua Lin, Chuzheng Wu, Li Ma
Summary: The shortest path problem in a fuzzy network involves determining the optimal path between specified source and destination vertices, utilizing fuzzy logic to handle uncertainties. The two main challenges in this context are calculating path length using fuzzy addition and comparing path lengths denoted by fuzzy parameters. Graded mean integration technique of triangular fuzzy numbers and genetic algorithm are commonly used to address these challenges.
COMPLEX & INTELLIGENT SYSTEMS
(2021)
