Article
Computer Science, Artificial Intelligence
Liang Mao, Shiliang Sun
Summary: This article extends VSGP to handle multiview data and demonstrates that the MVSGP model consistently outperforms single-view VSGP and state-of-the-art kernel-based multiview baselines for classification tasks on real-world datasets.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Automation & Control Systems
Haifeng Sima, Jing Wang, Ping Guo, Junding Sun, Hongmin Liu, Mingliang Xu, Youfeng Zou
Summary: The proposed method in this article, based on transfer learning of combined mid-level features, outperformed several state-of-the-art competitive algorithms for hyperspectral classification, demonstrating significant improvement in performance.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Computer Science, Artificial Intelligence
Fatemeh Alavi, Sattar Hashemi
Summary: The study introduces a bi-level learning paradigm for multiple kernel learning, which consists of kernel combination weight learning and self-paced learning stages that negotiate alternately, addressing the inadequacies of existing MKL methods on unreliable instances. Experimental results demonstrate the superiority and robustness of the proposed approach.
PATTERN RECOGNITION
(2022)
Article
Computer Science, Information Systems
Huibing Wang, Yang Wang, Zhao Zhang, Xianping Fu, Li Zhuo, Mingliang Xu, Meng Wang
Summary: In this paper, a general framework named KMSA is proposed for dimension reduction of multiview data, which can fully exploit information and differentiate between different views. With the co-regularized term and self-weighted learning, KMSA can learn appropriate weights for all views effectively.
IEEE TRANSACTIONS ON MULTIMEDIA
(2021)
Article
Computer Science, Artificial Intelligence
Chang-Dong Wang, Man-Sheng Chen, Ling Huang, Jian-Huang Lai, Philip S. Yu
Summary: The proposed multiview subspace clustering method, SMSCK, utilizes kernel learning and smoothness regularization to capture nonlinear relations between multiview data points and preserve the locality property of the original feature space. Theoretical analysis and experimental results demonstrate the effectiveness of the method in clustering multiview data.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Xu Yang, Cheng Deng, Zhiyuan Dang, Dacheng Tao
Summary: In this article, a novel multiview clustering model is proposed, which utilizes multiple autoencoder networks to embed multiview data into different latent spaces. A heterogeneous graph learning module is employed to adaptively fuse the latent representations, and intraview and interview collaborative learning are used to optimize the clustering results. Experimental results show that this method significantly outperforms other clustering approaches on multiple datasets.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Shirui Luo, Xiaochun Cao
Summary: This article introduces a technique that applies dual clustering to multiview subspace clustering, combining multiview learning and dual clustering segmentation to achieve better partitioning of data, and solving the problem using an alternating optimization scheme.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Biology
Luke Ternes, Mark Dane, Sean Gross, Marilyne Labrie, Gordon Mills, Joe Gray, Laura Heiser, Young Hwan Chang
Summary: The Multi-Encoder Variational AutoEncoder (ME-VAE) is a computational model that can control for multiple transformational features in single-cell imaging data, allowing for better separation of heterogeneous cell types and extraction of meaningful single-cell information. The ME-VAE improves analysis by enhancing phenotypic differences between cells and correlating with other analytic modalities. This new approach enables better feature extraction and image analysis methods, leading to advancements in understanding complex cell biology and improving medical outcomes and drug discovery.
COMMUNICATIONS BIOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Li Wang, Ren-Cang Li, Wen-Wei Lin
Summary: In this article, a subspace-based learning method using least squares as the fundamental basis is proposed. The method, called multiview orthonormalized partial least squares (MvOPLSs), learns a classifier over a common latent space shared by all views. Regularization techniques are leveraged to improve the performance of the method, and nonlinear transformations parameterized by deep networks are introduced for further enhancement. Extensive experiments show that the subspace-based learning for a common latent space is effective and its nonlinear extension can further boost performance, with one of the proposed methods achieving better results than all compared methods.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Mingjiang Liang, Shaoli Huang, Shirui Pan, Mingming Gong, Wei Liu
Summary: This paper proposes a multi-level weight-centric feature learning method to fully utilize the role of feature extractors in few-shot learning. Experimental results demonstrate that the proposed method outperforms its counterparts in low-shot classification tasks.
PATTERN RECOGNITION
(2022)
Article
Computer Science, Artificial Intelligence
Nan Zhang, Shiliang Sun
Summary: Multiview clustering is an important research topic, and incomplete views of data instances are common in real-world scenarios. To address this issue, we propose an effective incomplete multiview nonnegative representation learning framework that can handle incomplete multiview clustering in various situations and achieves better results compared to other state-of-the-art algorithms.
PATTERN RECOGNITION
(2022)
Article
Automation & Control Systems
Jie Wen, Zheng Zhang, Lunke Fei, Bob Zhang, Yong Xu, Zhao Zhang, Jinxing Li
Summary: Conventional multiview clustering methods fail when not all views of samples are available in practical applications. Incomplete multiview clustering (IMC) is developed to address this issue. Recent years have seen significant advances in IMC research, but there are still open problems to be solved.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2023)
Article
Environmental Sciences
Zhiyuan Li, Jiayi Guo, Yueting Zhang, Jie Li, Yirong Wu
Summary: In this work, a novel deep learning model called Ref-MFFDN is proposed for remote sensing image deblurring. By registering the reference image and the blurry image in the multi-level feature space, high-quality textures are transferred from the registered reference features to assist image deblurring, leading to improved deblurring performance.
Article
Computer Science, Artificial Intelligence
Wei Lv, Chao Zhang, Huaxiong Li, Xiuyi Jia, Chunlin Chen
Summary: This article addresses the issue of incomplete information in multiview clustering and proposes a method based on joint projection learning and tensor decomposition to solve this problem. The method projects high-dimensional features into a lower-dimensional space for compact feature learning and learns similarity graphs for instances of different views. These graphs are stacked into a third-order low-rank tensor to explore high-order correlations. The method also considers graph noise caused by missing samples and uses a tensor-decomposition-based graph filter for robust clustering.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Kaifeng Zou, Sylvain Faisan, Fabrice Heitz, Sebastien Valette
Summary: Two recent works have shown that modeling both high-level factors and their related features benefits the learning of disentangled representations with variational autoencoders (VAE). In this paper, a novel VAE-based approach is proposed, inspired by conditional VAE, which computes features deterministically using a neural network and allows for jointly estimating the parameters of the decoder, encoder, and feature network. The approach also improves the quality of generated images by using discrete latent variables and a two-step learning procedure, achieving better disentanglement performance and higher quality images compared to the two aforementioned works.
PATTERN RECOGNITION LETTERS
(2023)
Article
Infectious Diseases
Onder Ergonul, Merve Akyol, Cem Tanriover, Henning Tiemeier, Eskild Petersen, Nicola Petrosillo, Mehmet Gonen
Summary: The main factors affecting the CFR of COVID-19 include obesity rate, tuberculosis incidence, duration since the first death, and median age. Additionally, the test rate, hospital bed density, and rural population ratio also have an impact on the CFR.
CLINICAL MICROBIOLOGY AND INFECTION
(2021)
Article
Infectious Diseases
Volkan Korten, Deniz Gokengin, Gulhan Eren, Taner Yildirmak, Serap Gencer, Haluk Eraksoy, Dilara Inan, Figen Kaptan, Basak Dokuzoguz, Ilkay Karaoglan, Ayse Willke, Mehmet Gonen, Onder Ergonul
Summary: This study found that discontinuations of ART due to intolerance/toxicity and virologic failure decreased over time. InSTI-based regimens were less likely to be discontinued compared to PI- and NNRTI-based regimens.
AIDS RESEARCH AND THERAPY
(2021)
Article
Infectious Diseases
Onder Ergonul, Gizem Tokca, Siran Keske, Ebru Donmez, Bahar Madran, Azize Komur, Mehmet Gonen, Fusun Can
Summary: This paper describes the elimination of healthcare-associated Acinetobacter baumannii infections in a highly endemic region. The control of Acinetobacter baumannii outbreaks can be achieved by close follow-up supported by molecular techniques, strict application of infection control measures, and isolation of transferred patients.
INTERNATIONAL JOURNAL OF INFECTIOUS DISEASES
(2022)
Article
Infectious Diseases
Yusuf Mert Demirlenk, Lal Sude Gucer, Duygu Ucku, Cem Tanriover, Merve Akyol, Zeynepgul Kalay, Erinc Barcin, Rustu Emre Akcan, Fusun Can, Mehmet Gonen, Onder Ergonul
Summary: The addition of aminoglycosides to the existing treatment regimen for colistin- and carbapenem-resistant Klebsiella pneumoniae infections significantly reduces mortality, while treatment with tigecycline does not have a significant effect on mortality.
EUROPEAN JOURNAL OF CLINICAL MICROBIOLOGY & INFECTIOUS DISEASES
(2022)
Article
Infectious Diseases
Burcu Isler, Berna Ozer, Gule Cinar, Abdullah Tarik Aslan, Cansel Vatansever, Caitlin Falconer, Istar Dolapci, Funda Simsek, Necla Tulek, Hamiyet Demirkaya, Sirin Menekse, Halis Akalin, Ilker Inanc Balkan, Mehtap Aydin, Elif Tukenmez Tigen, Safiye Koculu Demir, Mahir Kapmaz, Siran Keske, Ozlem Dogan, Cigdem Arabaci, Serap Yagci, Gulsen Hazirolan, Veli Oguzalp Bakir, Mehmet Gonen, Mark D. Chatfield, Brian Forde, Nese Saltoglu, Alpay Azap, Ozlem Azap, Murat Akova, David L. Paterson, Fusun Can, Onder Ergonul
Summary: This study investigated carbapenem-resistant Klebsiella spp. bloodstream infections in Turkey and found important findings such as the prevalence of different types of carbapenemases and factors associated with mortality rates.
EUROPEAN JOURNAL OF CLINICAL MICROBIOLOGY & INFECTIOUS DISEASES
(2022)
Article
Endocrinology & Metabolism
Cem Sulu, Ayyuce Begum Bektas, Serdar Sahin, Emre Durcan, Zehra Kara, Ahmet Numan Demir, Hande Mefkure Ozkaya, Necmettin Tanriover, Nil Comunoglu, Osman Kizilkilic, Nurperi Gazioglu, Mehmet Gonen, Pinar Kadioglu
Summary: This study developed machine learning models to predict postoperative remission, remission at last visit, and resistance to somatostatin receptor ligands (SRL) in patients with acromegaly. The models identified clinical features associated with prognosis.
Article
Biochemical Research Methods
Ayyuce Begum Bektas, Cigdem Ak, Mehmet Gonen
Summary: With the increasing sizes of computational biology datasets, previous kernel-based machine learning algorithms have failed to provide satisfactory interpretability. To address this issue, we propose a fast and efficient multiple kernel learning algorithm that can extract significant information from genomic data. Our experiments demonstrate that the algorithm outperforms baseline methods while using only a small fraction of input features, and it has the potential to discover new biomarkers and therapeutic guidelines.
Article
Computer Science, Artificial Intelligence
Ruslan Khalitov, Tong Yu, Lei Cheng
Summary: This paper proposes a method to approximate a large square matrix with a product of sparse full-rank matrices, which is especially useful for scalable neural attention modeling. The experimental results demonstrate that our method gives a better approximation when the approximated matrix is sparse and high rank.
Article
Computer Science, Artificial Intelligence
Jing Zhang, YangLi-ao Geng, Wen Wang, Wenju Sun, Zhirong Yang, Qingyong Li
Summary: This paper investigates how to perform zero-shot learning with fewer seen samples. A Distribution and Gradient constrained Embedding Model (DGEM) is proposed to predict the visual prototypes for the given semantic vectors of seen classes. Experimental results show that DGEM outperforms other established methods when each seen class has only 1/5 sample(s).
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Automation & Control Systems
Arezou Rahimi, Mehmet Gonen
Summary: The study proposed a multitask MKL formulation with task clustering and a highly time-efficient solution approach. Experimental results showed that as the number of tasks and clusters increased, the forest formulation performed increasingly better in terms of computational performance.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Computer Science, Information Systems
Cigdem Ak, Alex D. Chitsazan, Mehmet Gonen, Ruth Etzioni, Aaron J. Grossberg
Summary: The impact of COVID-19 in the US varies across different areas, and this study examines the relationship between the risk of COVID-19, geographical location, and demographic features. The findings show that urbanicity and presidential vote margin are the most predictive factors for COVID-19 spread.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2022)
Article
Oncology
Milad Mokhtaridoost, Philipp G. Maass, Mehmet Gonen
Summary: Understanding the regulatory modules of miRNA-mRNA interactions in primary tumors is crucial for predicting tumor response to therapies and developing accurate treatment strategies. In this study, a computational pipeline was developed to extract tissue- and cohort-specific miRNA-mRNA regulatory modules from expression profiles of primary tumors. The model effectively identified cohort-specific and tissue-specific regulatory modules, and demonstrated its ability to determine cancer-related miRNAs and extract tissue-specific information.
Article
Computer Science, Theory & Methods
Zhirong Yang, Yuwei Chen, Denis Sedov, Samuel Kaski, Jukka Corander
Summary: Neighbor Embedding (NE) is an effective principle for data visualization, but current methods may hide large-scale patterns. To address this, we propose a new cluster visualization method based on the NE principle and present a family of NE methods that can better display clusters.
STATISTICS AND COMPUTING
(2023)
Proceedings Paper
Computer Science, Information Systems
Marzieh Soleimanpoor, Milad Mokhtaridoost, Mehmet Gonen
Summary: Standard machine learning algorithms have limited knowledge extraction capability in discriminating cancer stages based on genomic characterizations. We implemented a kernel-based neural network framework that integrates pathways and gene expression data using multiple kernels and discriminates early- and late-stages of cancers. Our method obtained better or comparable predictive performance against existing classification algorithms.
MACHINE LEARNING, OPTIMIZATION, AND DATA SCIENCE, LOD 2022, PT I
(2023)
Proceedings Paper
Computer Science, Information Systems
Emre Atan, Ali Duymaz, Funda Sarisozen, Ugur Aydin, Murat Koras, Baris Akgun, Mehmet Gonen
Summary: We constructed a financial network based on customer relationships and money transactions. The study aims to identify profitable customers and analyze the impact of financial deterioration on related customers. The findings show that the top 30% customers in terms of centrality have five times higher profitability and the variables created contribute to the model used by the bank.
MACHINE LEARNING, OPTIMIZATION, AND DATA SCIENCE, LOD 2022, PT I
(2023)
Article
Computer Science, Artificial Intelligence
Rui Lv, Dingheng Wang, Jiangbin Zheng, Zhao-Xu Yang
Summary: In this paper, the authors investigate tensor decomposition for neural network compression. They analyze the convergence and precision of tensor mapping theory, validate the rationality of tensor mapping and its superiority over traditional tensor approximation based on the Lottery Ticket Hypothesis. They propose an efficient method called 3D-KCPNet to compress 3D convolutional neural networks using the Kronecker canonical polyadic (KCP) tensor decomposition. Experimental results show that 3D-KCPNet achieves higher accuracy compared to the original baseline model and the corresponding tensor approximation model.
Article
Computer Science, Artificial Intelligence
Xiangkun He, Zhongxu Hu, Haohan Yang, Chen Lv
Summary: In this paper, a novel constrained multi-objective reinforcement learning algorithm is proposed for personalized end-to-end robotic control with continuous actions. The approach trains a single model using constraint design and a comprehensive index to achieve optimal policies based on user-specified preferences.
Article
Computer Science, Artificial Intelligence
Zhijian Zhuo, Bilian Chen, Shenbao Yu, Langcai Cao
Summary: In this paper, a novel method called Expansion with Contraction Method for Overlapping Community Detection (ECOCD) is proposed, which utilizes non-negative matrix factorization to obtain disjoint communities and applies expansion and contraction processes to adjust the degree of overlap. ECOCD is applicable to various networks with different properties and achieves high-quality overlapping community detection.
Article
Computer Science, Artificial Intelligence
Yizhe Zhu, Chunhui Zhang, Jialin Gao, Xin Sun, Zihan Rui, Xi Zhou
Summary: In this work, the authors propose a Contrastive Spatio-Temporal Distilling (CSTD) approach to improve the detection of high-compressed deepfake videos. The approach leverages spatial-frequency cues and temporal-contrastive alignment to fully exploit spatiotemporal inconsistency information.
Review
Computer Science, Artificial Intelligence
Laijin Meng, Xinghao Jiang, Tanfeng Sun
Summary: This paper provides a review of coverless steganographic algorithms, including the development process, known contributions, and general issues in image and video algorithms. It also discusses the security of coverless steganography from theoretical analysis to actual investigation for the first time.
Article
Computer Science, Artificial Intelligence
Yajie Bao, Tianwei Xing, Xun Chen
Summary: Visual question answering requires processing multi-modal information and effective reasoning. Neural-symbolic learning is a promising method, but current approaches lack uncertainty handling and can only provide a single answer. To address this, we propose a confidence based neural-symbolic approach that evaluates NN inferences and conducts reasoning based on confidence.
Article
Computer Science, Artificial Intelligence
Anh H. Vo, Bao T. Nguyen
Summary: Interior style classification is an interesting problem with potential applications in both commercial and academic domains. This project proposes a method named ISC-DeIT, which combines data-efficient image transformer architectures and knowledge distillation, to address the interior style classification problem. Experimental results demonstrate a significant improvement in predictive accuracy compared to other state-of-the-art methods.
Article
Computer Science, Artificial Intelligence
Shashank Kotyan, Danilo Vasconcellos Vargas
Summary: This article introduces a novel augmentation technique called Dynamic Scanning Augmentation to improve the accuracy and robustness of Vision Transformer (ViT). The technique leverages dynamic input sequences to adaptively focus on different patches, resulting in significant changes in ViT's attention mechanism. Experimental results demonstrate that Dynamic Scanning Augmentation outperforms ViT in terms of both robustness to adversarial attacks and accuracy against natural images.
Article
Computer Science, Artificial Intelligence
Hiba Alqasir, Damien Muselet, Christophe Ducottet
Summary: The article proposes a solution to improve the learning process of a classification network by providing shape priors, reducing the need for annotated data. The solution is tested on cross-domain digit classification tasks and a video surveillance application.
Article
Computer Science, Artificial Intelligence
Dexiu Ma, Mei Liu, Mingsheng Shang
Summary: This paper proposes a method using neural dynamics solvers to solve infinity-norm optimization problems. Two improved solvers are constructed and their effectiveness and superiority are demonstrated through theoretical analysis and simulation experiments.
Article
Computer Science, Artificial Intelligence
Francesco Gregoretti, Giovanni Pezzulo, Domenico Maisto
Summary: Active Inference is a computational framework that uses probabilistic inference and variational free energy minimization to describe perception, planning, and action. cpp-AIF is a header-only C++ library that provides a powerful tool for implementing Active Inference for Partially Observable Markov Decision Processes through multi-core computing. It is cross-platform and improves performance, memory management, and usability compared to existing software.
Article
Computer Science, Artificial Intelligence
Zelin Ying, Dawei Cheng, Cen Chen, Xiang Li, Peng Zhu, Yifeng Luo, Yuqi Liang
Summary: This paper proposes a novel stock market trends prediction framework called SMART, which includes a self-supervised stock technical data sequence embedding model S3E. By training with multiple self-supervised auxiliary tasks, the model encodes stock technical data sequences into embeddings and uses the learned sequence embeddings for predicting stock market trends. Extensive experiments on China A-Shares market and NASDAQ market prove the high effectiveness of our model in stock market trends prediction, and its effectiveness is further validated in real-world applications in a leading financial service provider in China.
Article
Computer Science, Artificial Intelligence
Hao Li, Hao Jiang, Dongsheng Ye, Qiang Wang, Liang Du, Yuanyuan Zeng, Liu Yuan, Yingxue Wang, C. Chen
Summary: DHGAT1, a dynamic hyperbolic graph attention network, utilizes hyperbolic metric properties to embed dynamic graphs. It employs a spatiotemporal self-attention mechanism and weighted node representations, resulting in excellent performance in link prediction tasks.
Article
Computer Science, Artificial Intelligence
Jiehui Huang, Zhenchao Tang, Xuedong He, Jun Zhou, Defeng Zhou, Calvin Yu-Chian Chen
Summary: This study proposes a progressive learning multi-scale feature blending model for image deraining tasks. The model utilizes detail dilation and texture extraction to improve the restoration of rainy images. Experimental results show that the model achieves near state-of-the-art performance in rain removal tasks and exhibits better rain removal realism.
Article
Computer Science, Artificial Intelligence
Lizhi Liu, Zilin Gao, Yinhe Wang, Yongfu Li
Summary: This paper proposes a novel discrete-time interconnected model for depicting complex dynamical networks. The model consists of nodes and edges subsystems, which consider the dynamic characteristic of both nodes and edges. By designing control strategies and coupling modes, the stabilization and synchronization of the network are achieved. Simulation results demonstrate the effectiveness of the proposed methods.