Article
Computer Science, Artificial Intelligence
Prasad Bhavana, Vineet Padmanabhan
Summary: The paper introduces a divide and conquer technique based on a two stage factorization process to address memory limitations and computational efficiency in matrix factorization tasks, playing a crucial role in industrial applications.
APPLIED INTELLIGENCE
(2021)
Article
Computer Science, Artificial Intelligence
Guojie Song, Liang Zhang, Ziyao Li, Yi Li
Summary: This paper introduces a new network embedding algorithm called SepNE, which learns representations for different subsets of nodes independently in a separated process, thereby improving the scalability of large-scale networks. By preserving both local and global information in the objective function, this algorithm is able to capture more information and leverage high-order proximities in large networks using several methods.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Mathematics, Applied
Hong Zhu, Michael K. Ng, Guang-Jing Song
Summary: A new approximate method for solving nonnegative low-rank matrix approximation problem is developed in this study. It involves alternately projecting onto fixed-rank matrix manifold and nonnegative matrix manifold to ensure convergence, with numerical results demonstrating its performance.
JOURNAL OF SCIENTIFIC COMPUTING
(2021)
Article
Mathematics, Applied
Nicoletta Del Buono, Flavia Esposito, Laura Selicato, Rafal Zdunek
Summary: Learning approaches rely on hyperparameters that impact algorithm performance and affect data knowledge extraction. Nonnegative Matrix Factorization (NMF) has gained interest as a learning algorithm for capturing latent information in large datasets while maintaining feature properties. Tuning the penalty hyperparameters in NMF is an open challenge. This study proposes a bi-level optimization framework for addressing the penalty hyperparameters problem in NMF and introduces a novel algorithm called Alternating Bi-level (AltBi) that incorporates hyperparameters tuning into NMF updates. The study investigates the existence and convergence of numerical solutions under certain assumptions and presents numerical experiments.
APPLIED MATHEMATICS AND COMPUTATION
(2023)
Article
Ecology
L. M. Rivera-Munoz, A. F. Giraldo-Forero, J. D. Martinez-Vargas
Summary: According to the WHO, pollution is a global public health issue. In Colombia, low-cost strategies using wireless sensor networks (WSNs) have been implemented for air quality monitoring. However, data missing is a common problem in WSNs due to environmental and location conditions. This study proposes a novel deep matrix factorization technique to estimate missing particulate matter data in a WSN, which outperforms standard matrix factorization and other variations of the model.
ECOLOGICAL INFORMATICS
(2022)
Article
Computer Science, Artificial Intelligence
Yun Liu, Shujuan Ji, Qiang Fu, Dickson K. W. Chiu, Maoguo Gong
Summary: The proposed Dual Semantic Preserving Hashing (DSPH) method addresses the challenges of semantic information utilization and discriminative hash code learning in cross-modal hashing by leveraging matrix factorization and discrete optimization strategy.
Article
Mathematical & Computational Biology
Wenlong Ma, Siyuan Chen, Yuhong Qi, Minggui Song, Jingjing Zhai, Ting Zhang, Shang Xie, Guifeng Wang, Chuang Ma
Summary: In this study, we developed easyMF, a web platform that utilizes matrix factorization algorithms for functional gene discovery from large-scale transcriptome data. Compared with existing software, easyMF offers greater functionality, flexibility, and ease of use. The platform is equipped with user-friendly graphic user interfaces and supports various analyses, including transcriptome analysis, multiple-scenario matrix factorization analysis, and multiple-way gene discovery. We applied easyMF to maize RNA-Seq datasets and successfully identified numerous seed-specific genes. Additionally, easyMF outperformed other systems in gene prioritization.
INTERDISCIPLINARY SCIENCES-COMPUTATIONAL LIFE SCIENCES
(2022)
Article
Computer Science, Information Systems
Yun Liu, Shujuan Ji, Qiang Fu, Dickson K. W. Chiu
Summary: In this paper, an efficient Semantic-consistency Asymmetric Matrix Factorization Hashing (SAMFH) method is proposed to address the challenge of effectively exploiting semantic information for learning discriminative hash codes in cross-modal retrieval tasks.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Mathematics, Applied
Dongping Li, Xiuying Zhang, Renyun Liu
Summary: This paper discusses numerical integration for large-scale systems of stiff Riccati differential equations using exponential Rosenbrock-type integrators, addressing implementation issues and utilizing low-rank approximations based on high quality numerical algebra codes. Numerical comparisons demonstrate the high accuracy and efficiency of exponential integrators for solving large-scale systems of stiff Riccati differential equations.
JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS
(2021)
Article
Computer Science, Software Engineering
Zikun Deng, Di Weng, Yuxuan Liang, Jie Bao, Yu Zheng, Tobias Schreck, Mingliang Xu, Yingcai Wu
Summary: The article introduces a visual analytics system called VisCas, which aims to mine and interpret cascading patterns in urban contexts. The system combines an inference model with interactive visualizations to empower analysts to infer and interpret latent cascading patterns. It addresses challenges in generalized pattern inference, implicit influence visualization, and multifaceted cascade analysis. The effectiveness of VisCas is demonstrated through case studies on real-world traffic congestion and air pollution datasets.
IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS
(2022)
Article
Computer Science, Artificial Intelligence
Tao Yao, Yiru Li, Weili Guan, Gang Wang, Ying Li, Lianshan Yan, Qi Tian
Summary: Cross-media hashing encodes data points from different modalities into a common Hamming space, and has been successfully applied to large-scale multimedia retrieval. However, existing methods neglect the potential inconsistency among different modalities, which may undermine retrieval accuracy. To address this problem, we propose a novel unsupervised hashing model, DRMFH, which formulates the consistency and inconsistency across different modalities into a matrix factorization based model.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Automation & Control Systems
Chencheng Cai, Rong Chen, Han Xiao
Summary: This study addresses the problem of matrix approximation and denoising caused by the Kronecker product decomposition. A method called Kronecker product approximation (KoPA) is proposed, which approximates a given matrix by the sum of several Kronecker products of matrices. By using extended information criteria to select the appropriate configuration, the proposed method is able to select the true configuration with high probability and outperforms the low rank approximations.
JOURNAL OF MACHINE LEARNING RESEARCH
(2022)
Article
Chemistry, Analytical
Murad Tukan, Alaa Maalouf, Matan Weksler, Dan Feldman
Summary: The study introduces an algorithm for compressing neural networks that utilizes modern techniques in computational geometry to approximate lp instead of k-rank l2 for effective compression. Experimental results confirm the practicality and theoretical advantage of this method in compressing networks such as BERT, DistilBERT, XLNet, and RoBERTa on the GLUE benchmark.
Article
Computer Science, Information Systems
Fan Yang, Qiaoxi Zhang, Fumin Ma, Xiaojian Ding, Yufeng Liu, Deyu Tong
Summary: This paper proposes an efficient discrete cross-modal hashing method that incorporates an asymmetric model, semantic supervised intersection scheme, category correlations embedding, optimization strategy, and linear projection to improve retrieval performance and effectiveness.
INFORMATION SCIENCES
(2023)
Article
Engineering, Electrical & Electronic
Shuang Xu, Jiangshe Zhang, Chunxia Zhang
Summary: This paper proposes a new method for hyperspectral image denoising, called Hyper-Laplacian spectral-spatial total variation (HTV), and designs two low-rank models. Experimental results demonstrate the superiority of the HTV method over traditional TV regularization methods and other commonly used hyperspectral image denoising algorithms.
Article
Computer Science, Artificial Intelligence
Marion Neumann, Roman Garnett, Christian Bauckhage, Kristian Kersting
Article
Computer Science, Information Systems
Shanshan Zhang, Dominik A. Klein, Christian Bauckhage, Armin B. Cremers
MULTIMEDIA TOOLS AND APPLICATIONS
(2016)
Article
Neurosciences
Marlene Hoehne, Amirhossein Jahanbekam, Christian Bauckhage, Nikolai Axmacher, Juergen Fell
Article
Chemistry, Analytical
Matthias Alfeld, Mirwaes Wahabzada, Christian Bauckhage, Kristian Kersting, Geert van der Snickt, Petria Noble, Koen Janssens, Gerd Wellenreuther, Gerald Falkenberg
MICROCHEMICAL JOURNAL
(2017)
Article
Neurosciences
Marlene Derner, Amirhossein Jahanbekam, Christian Bauckhage, Nikolai Axmacher, Juergen Fell
EUROPEAN JOURNAL OF NEUROSCIENCE
(2018)
Article
Biology
Maurice Guender, Facundo R. Ispizua Yamati, Jana Kierdorf, Ribana Roscher, Anne-Katrin Mahlein, Christian Bauckhage
Summary: In this work, a hands-on workflow for the automatized temporal and spatial identification and individualization of crop images from UAVs is presented, improving the analysis and interpretation of UAV data in agriculture significantly. The results show that the approach has similar accuracy to more complex deep learning-based recognition techniques and can automate the processing of large datasets.
Article
Computer Science, Information Systems
Maurice Gunder, Nico Piatkowski, Laura Von Rueden, Rafet Sifa, Christian Bauckhage
Summary: Efficient and economical usage of agricultural land is increasingly important in the face of climate change and resource scarcity. Intercropping of various plant species is recommended to avoid the disadvantages of monocropping, but it poses challenges due to the need for balanced planting schedules. The proposed flexible optimization method aims to address these challenges by combining evolutionary algorithms with a hierarchical loss function and adaptive mutation rate, leading to faster and better solutions for a sustainable crop harvesting season.
Proceedings Paper
Computer Science, Artificial Intelligence
C. Bauckhage, R. Ramamurthy, R. Sifa
ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING, ICANN 2020, PT II
(2020)
Article
Computer Science, Artificial Intelligence
Rafet Sifa, Raheel Yawar, Rajkumar Ramamurthy, Christian Bauckhage, Kristian Kersting
KUNSTLICHE INTELLIGENZ
(2020)
Proceedings Paper
Computer Science, Artificial Intelligence
Christian Bauckhage, Rafet Sifa, Tiansi Dong
ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING - ICANN 2019: WORKSHOP AND SPECIAL SESSIONS
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
R. Ramamurthy, C. Bauckhage, R. Sifa, S. Wrobel
ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING - ICANN 2018, PT III
(2018)
Proceedings Paper
Computer Science, Artificial Intelligence
Rafet Sifa, Daniel Paurat, Daniel Trabold, Christian Bauckhage
ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING - ICANN 2018, PT III
(2018)
Proceedings Paper
Computer Science, Artificial Intelligence
Benjamin Wulff, Jannis Schuecker, Christian Bauckhage
ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING - ICANN 2018, PT III
(2018)
Proceedings Paper
Computer Science, Artificial Intelligence
Christian Bauckhage
ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING - ICANN 2017, PT I
(2017)
Proceedings Paper
Computer Science, Artificial Intelligence
Rajkumar Ramamurthy, Christian Bauckhage, Krisztian Buza, Stefan Wrobel
ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING, PT II
(2017)