Article
Computer Science, Artificial Intelligence
Jaemin Yoo, Junghun Kim, Hoyoung Yoon, Geonsoo Kim, Changwon Jang, U. Kang
Summary: In this work, the authors propose a novel graph-based positive-unlabeled (PU) learning method called GRAB, which requires no class prior. They further generalize GRAB to multi-positive unlabeled (MPU) learning. Experimental results demonstrate that GRAB achieves state-of-the-art performance on several real-world datasets, even without the true prior information given to competitors.
KNOWLEDGE AND INFORMATION SYSTEMS
(2022)
Article
Geochemistry & Geophysics
Xiao Ling, Rongjun Qin
Summary: This article proposes a novel and efficient texture mapping framework that allows the use of multiple texture views per face while achieving global color consistency. The proposed method utilizes a loopy belief propagation algorithm to perform efficient and global-level probabilistic inferences for candidate views per face, enabling face-level multiview texture fusion and blending.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Multidisciplinary Sciences
M. E. J. Newman
Summary: Networks and network computations have become essential for analyzing complex systems. Message passing methods, which involve the propagation of information between network nodes, are commonly used for calculating quantities on nodes. This perspective article discusses the application of message passing methods, provides examples and applications, and explores the connection between message passing and phase transitions in networks. It also discusses the limitations of message passing methods and describes recent methods that address these limitations.
PROCEEDINGS OF THE ROYAL SOCIETY A-MATHEMATICAL PHYSICAL AND ENGINEERING SCIENCES
(2023)
Article
Physics, Fluids & Plasmas
Peter Mann, Simon Dobson
Summary: In this paper, the authors propose a new approach using belief propagation and edge-disjoint motif covers to study bond percolation on random and real world networks. They derive exact message passing expressions for cliques and chordless cycles and find good agreement with Monte Carlo simulation. This approach offers a simple and substantial improvement on traditional message passing methods.
Article
Computer Science, Artificial Intelligence
Daqi Liu, Miroslaw Bober, Josef Kittler
Summary: Scene graph generation aims to interpret an input image by explicitly modelling the objects contained therein and their relationships. Existing methods predominantly use message passing neural network models to solve this problem. However, in these models, the variational distributions often ignore the structural dependencies among output variables, and most scoring functions only consider pairwise dependencies, which may lead to inconsistent interpretations. This article proposes a novel neural belief propagation method that replaces the traditional mean field approximation with a structural Bethe approximation. Higher-order dependencies among three or more output variables are also incorporated into the relevant scoring function to find a better bias-variance trade-off. The proposed method achieves state-of-the-art performance on various popular scene graph generation benchmarks.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Engineering, Electrical & Electronic
Wenbo Ding, Dongsheng Wang, Yang Li
Summary: This paper proposes a structured variational inference algorithm to address the issue of significant multipath components in RF sensing information obtained in household environments. By using multipath components wisely, the algorithm improves sensing accuracy and enables joint estimation of reflector and target states. It utilizes belief propagation algorithm to approximate the discrete association variables in the multi-frame association problem, and mean-field variation to approximate the hierarchical measurement model. Simulation experiments demonstrate that the proposed algorithm achieves better estimation accuracy than traditional methods.
DIGITAL SIGNAL PROCESSING
(2023)
Article
Computer Science, Artificial Intelligence
Lianmeng Jiao, Feng Wang, Zhun-ga Liu, Quan Pan
Summary: In this study, a transfer learning-based evidential clustering algorithm (TECM) is proposed to address the issue of insufficient or contaminated data on clustering performance. The TECM algorithm integrates knowledge learned from a source domain with the data in a target domain to cluster the target data, demonstrating its effectiveness compared to other representative multitask or transfer-clustering algorithms.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Multidisciplinary Sciences
Chen Liang, Xiaofeng Wang, Lei Lu, Pengfei Niu
Summary: The study presents an analysis method for solution space structure based on planting strategy and belief propagation, which can effectively analyze the evolution process of 3-SAT problem solution space and satisfiability phase transition, confirming the validity of the research.
Article
Engineering, Electrical & Electronic
Sang Hyun Lee, Mintae Kim, Hunmin Shin, Inkyu Lee
Summary: This article studies the energy efficient management of two-tier heterogeneous cellular networks, proposing a distributed user association algorithm that maximizes network-wide energy efficiency. By turning off BSs that support only a small number of users and offloading serving users to adjacent active BSs, the network-wide energy consumption is minimized while the sum throughput is maximized. Introducing a new approach based on a message-passing framework and deriving a distributed load balancing algorithm, the proposed method provides an efficient solution with reduced computational complexity compared to existing schemes.
IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Zuo-wei Zhang, Zhe Liu, Arnaud Martin, Zhun-ga Liu, Kuang Zhou
Summary: The paper introduces a dynamic evidential clustering algorithm (DEC) to reduce the computational burden of existing methods. By minimizing an FCM-like objective function, it obtains the support levels of the real singleton clusters to which query objects belong, and assigns the query objects to outlier, precise, or imprecise clusters based on conflicts between support levels. The proposed method effectively extends the application of evidential clustering, especially in big data scenarios, by reducing complexity and testing its effectiveness with artificial and real datasets.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Hao Luo, Qianli Zhou, Zhen Li, Yong Deng
Summary: Dempster-Shafer Theory (DST) is widely used in decision making and information fusion, but its exponential computational complexity limits real-time application. This paper proposes a VQLS-based method to conduct DST operations with lower error level and fewer quantum resources, suitable for the NISQ era.
INFORMATION FUSION
(2024)
Article
History & Philosophy Of Science
Robert Weston Siscoe
Summary: Traditional epistemologists believe that rational requirements on belief are the most important norms, but revolutionary epistemologists argue that norms on credences are more fundamental and determinative. However, measuring accuracy poses a challenge for the revolutionary epistemologists.
Article
Chemistry, Multidisciplinary
Amedeo Buonanno, Antonio Nogarotto, Giuseppe Cacace, Giovanni Di Gennaro, Francesco A. N. Palmieri, Maria Valenti, Giorgio Graditi
Summary: The work investigates an Information Fusion architecture based on a Factor Graph in Reduced Normal Form, allowing for a completely probabilistic framework to describe fusion. This architecture is flexible, extendable, and robust, making it suitable for scenarios where the signal can be wrongly received or completely missing.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Information Systems
Bolun Zhang, Guangen Gao, Yongxin Gao
Summary: Researchers propose an augmented-state BP algorithm to alleviate the effect of loops in mobile agent networks for cooperative localization problems, which improves localization performance by reducing the number of loops in the factor graph through state augmentation.
Article
Engineering, Multidisciplinary
Wei Wang, Michael K. Ng
Summary: This paper presents a novel inverse filtering method that combines a learned dictionary and an inverse filter to deconvolute observed images with unknown point spread functions. The alternating direction method of multipliers is employed to solve the optimization problem, and experimental results show that the proposed method outperforms other testing methods for various images.
APPLIED MATHEMATICAL MODELLING
(2021)
Article
Computer Science, Artificial Intelligence
Fabio Gagliardi Cozman, Denis Deratani Maua
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2019)
Article
Computer Science, Artificial Intelligence
Denis Deratani Maua, Fabio Gagliardi Cozman
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2020)
Article
Computer Science, Artificial Intelligence
Fabio Gagliardi Cozman, Denis Deratani Maua
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2020)
Article
Computer Science, Artificial Intelligence
Fabio Gagliardi Cozman
Summary: The paper examines various concepts of independence associated with different types of sets of probabilities and evaluates them based on the graphoid properties they satisfy. The analysis focuses on sets of probability measures, discussing versions of epistemic, confirmational, and type-5 independence based on regular conditioning, as well as complete and strong independence. Analogous concepts of independence for sets of full conditional probabilities, sets of lexicographic probabilities, and sets of desirable gambles are also discussed.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2021)
Article
Computer Science, Artificial Intelligence
Fabio Gagliardi Cozman, Hugo Neri Munhoz
Summary: This paper discusses how to use theoretical insights and practical tools from knowledge representation and reasoning to enhance machine learning, and when it is worthwhile to do so. It focuses on the knowledge representation and reasoning side of knowledge-enhanced machine learning, presenting case studies including probabilistic languages, explanations for embeddings, and an explainable version of the Winograd Challenge.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2021)
Article
Neurosciences
Yu Chen, Jaime S. Ide, Clara S. Li, Shefali Chaudhary, Thang M. Le, Wuyi Wang, Simon Zhornitsky, Sheng Zhang, Chiang-Shan R. Li
Summary: In this study, voxel-based morphometry analysis was used to investigate the cerebral volumetric correlates of impulsivity in a large sample of adolescent subjects. The results showed that different dimensions of impulsivity were associated with varying patterns of gray matter volumes, and there were also sex differences and genetic influences on these correlations.
HUMAN BRAIN MAPPING
(2022)
Article
Engineering, Marine
Gustavo A. Bisinotto, Joao V. Sparano, Alexandre N. Simos, Fabio G. Cozman, Marcos D. Ferreira, Eduardo A. Tannuri
Summary: This paper presents a study on motion-based wave estimation using neural networks. The performance of the data-driven inference system with motion data from different draft conditions is evaluated. The results show good agreement between estimations and reference values, but there are limitations when it comes to less frequent sea conditions. Additionally, the analysis of the estimation models' robustness with respect to draft variation demonstrates small average deviations compared to expected values.
Proceedings Paper
Computer Science, Artificial Intelligence
Marlon S. Mathias, Wesley P. de Almeida, Jefferson F. Coelho, Lucas P. de Freitas, Felipe M. Moreno, Caio F. D. Netto, Fabio G. Cozman, Anna Helena Reali Costa, Eduardo A. Tannuri, Edson S. Gomi, Marcelo Dottori
Summary: In this paper, we implement a Physics-Informed Neural Network (PINN) to solve the two-dimensional Burgers equations. By comparing PINNs trained with different amounts of governing equation evaluation points and known solution points, we find that models trained with the governing equations exhibit better overall observance of the underlying physics. We also investigate the impact of changing the number of each type of point on the resulting models. Finally, we argue that adding the governing equations during training can improve the overall performance of the model, particularly when the number of known solution points is limited.
INTELLIGENT SYSTEMS, PT II
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Paulo Pirozelli, Anarosa A. F. Brandao, Sarajane M. Peres, Fabio G. Cozman
Summary: This paper explores dual system architectures to improve the quality of answers generated by QA systems by filtering questions. Two experiments show that using classification and regression models to filter questions can enhance the accuracy of the answers.
INTELLIGENT SYSTEMS, PT II
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Marcos Menon Jose, Marcelo Archanjo Jose, Denis Deratani Maua, Fabio Gagliardi Cozman
Summary: Deep learning transformers have greatly improved systems that automatically answer questions in natural language. This study proposes an architecture that integrates different modules to answer two types of queries and validates the modular question answering strategy through experiments conducted in the Portuguese language.
COMPUTATIONAL PROCESSING OF THE PORTUGUESE LANGUAGE, PROPOR 2022
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Juliana Cesaro, Fabio Gagliardi Cozman
MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES, ECML PKDD 2019, PT I
(2020)
Article
Computer Science, Artificial Intelligence
Denis Deratani Maua, Fabio Gagliardi Cozman
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2020)
Article
Computer Science, Information Systems
Jose Amendola, Lucas S. Miura, Anna H. Reali Costa, Fabio G. Cozman, Eduardo Aoun Tannuri
Article
Computer Science, Artificial Intelligence
Hugo Neri, Fabio Cozman
Proceedings Paper
Computer Science, Artificial Intelligence
Rodrigo Monteiro de Aquino, Fabio Gagliardi Cozman
2019 IEEE SECOND INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND KNOWLEDGE ENGINEERING (AIKE)
(2019)
Article
Computer Science, Artificial Intelligence
Timotheus Kampik, Kristijonas Cyras, Jose Ruiz Alarcon
Summary: This paper presents a formal approach to explaining changes in inference in Quantitative Bipolar Argumentation Frameworks (QBAFs). The approach traces the causes of strength inconsistencies and provides explanations for them.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Xiangnan Zhou, Longchun Wang, Qingguo Li
Summary: This paper aims to establish a closer connection between domain theory and Formal Concept Analysis (FCA) by introducing the concept of an optimized concept for a formal context. With the utilization of optimized concepts, it is demonstrated that the class of formal contexts directly corresponds to algebraic domains. Additionally, two subclasses of formal contexts are identified to characterize algebraic L-domains and Scott domains. An application is presented to address the open problem of reconstructing bounded complete continuous domains using attribute continuous contexts, and the presentation of algebraic domains is extended to a categorical equivalence.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Sihan Wang, Zhong Yuan, Chuan Luo, Hongmei Chen, Dezhong Peng
Summary: Anomaly detection is widely used in various fields, but most current methods only work for specific data and ignore uncertain information such as fuzziness. This paper proposes an anomaly detection algorithm based on fuzzy rough entropy, which effectively addresses the similarity between high-dimensional objects using distance and correlation measures. The algorithm is compared and analyzed with mainstream anomaly detection algorithms on publicly available datasets, showing superior performance and flexibility.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Mario Alviano, Francesco Bartoli, Marco Botta, Roberto Esposito, Laura Giordano, Daniele Theseider Dupre
Summary: This paper investigates the relationships between a multipreferential semantics in defeasible reasoning and a multilayer neural network model. Weighted knowledge bases are considered for a simple description logic with typicality under a concept-wise multipreference semantics. The semantics is used to interpret MultiLayer Perceptrons (MLPs) preferentially. Model checking and entailment based approach are employed in verifying conditional properties of MLPs.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Bazin Alexandre, Galasso Jessie, Kahn Giacomo
Summary: Formal concept analysis is a mathematical framework that represents the information in binary object-attribute datasets using a lattice of formal concepts. It has been extended to handle more complex data types, such as relational data and n-ary relations. This paper presents a framework for polyadic relational concept analysis, which extends relational concept analysis to handle relational datasets consisting of n-ary relations.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Ander Gray, Marcelo Forets, Christian Schilling, Scott Ferson, Luis Benet
Summary: The presented method combines reachability analysis and probability bounds analysis to handle imprecisely known random variables. It can rigorously compute the temporal evolution of p-boxes and provide interval probabilities for formal verification problems. The method does not impose strict constraints on the input probability distribution or p-box and can handle multivariate p-boxes with a consonant approximation method.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Laszlo Csato
Summary: This paper studies a special type of incomplete pairwise comparison matrices and proposes a new method to determine the missing elements without violating the ordinal property.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
