Article
Computer Science, Information Systems
Jinchao Zheng, Rubin Wang, Wanzeng Kong, Jianhai Zhang
Summary: This paper studies the operation way of neural coding using the principles of minimum mutual information and maximum entropy. It explores the relationship between neural information and energy utilization, proposes the concept of energy neural coding, and provides a theoretical basis for understanding brain activity and brain-like artificial intelligence.
INFORMATION SCIENCES
(2022)
Article
Quantum Science & Technology
Xiaomin Liu, Zhengjun Xi, Heng Fan
Summary: This paper investigates the problem of classical Slepian-Wolf coding with quantum side information. Achievable rate regions are derived using quantum Feinstein's lemma and the measure compression theorem, and the results are extended to the case of multiple classical-quantum sources.
QUANTUM INFORMATION PROCESSING
(2023)
Article
Physics, Multidisciplinary
Garin Newcomb, Khalid Sayood
Summary: An important step in genome annotation is identifying protein-coding regions, and we propose signals based on average mutual information for accurate identification of coding and noncoding sequences. These signals are robust across species, phyla, and kingdom, making them suitable for species-agnostic genome annotation algorithms to identify protein-coding regions, which can aid in gene identification.
Article
Thermodynamics
Shengyuan Zhong, Jun Zhao, Wenjia Li, Hao Li, Shuai Deng, Yang Li, Sajjad Hussain, Xiaoyuan Wang, Jiebei Zhu
Summary: The study used mutual information to quantitatively analyze energy consumption fluctuations in building energy systems due to differences in information interaction. It found that the second control strategy performed the best when information differences were minimal.
Article
Computer Science, Artificial Intelligence
Xiaoqiang Yan, Yiqiao Mao, Yangdong Ye, Hui Yu
Summary: Cross-modal clustering aims to improve clustering accuracy by exploiting correlations across modalities. We propose a novel deep correlated information bottleneck (DCIB) method to tackle the challenges of capturing correlations and eliminating modality-private information. DCIB treats the CMC task as a two-stage data compression procedure, preserving correlations while eliminating modality-private information. Experimental results on four cross-modal datasets validate the superiority of DCIB.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Biochemical Research Methods
Antoine Passemiers, Yves Moreau, Daniele Raimondi
Summary: This article presents a novel method called PORTIA for inferring gene regulatory networks (GRNs). The method is based on robust precision matrix estimation and is shown to outperform state-of-the-art methods in terms of speed while still maintaining good accuracy. The authors extensively validated PORTIA using benchmark datasets and propose a new scoring metric based on graph-theoretical concepts.
Article
Computer Science, Artificial Intelligence
Dafeng Wang, Hongbo Liu, Naiyao Wang, Yiyang Wang, Hua Wang, Sean McLoone
Summary: In this paper, a novel Sequence Entropy Energy-based Model (SEEM) is proposed to address the limitations of current trajectory prediction models. SEEM achieves diversity in candidate trajectory generation by optimizing sequence entropy, and improves accuracy and stability through probability distribution clipping mechanism and energy network. Experimental results demonstrate that SEEM outperforms the state-of-the-art approaches in terms of diversity, accuracy, and stability of pedestrian trajectory prediction.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Martin Hofmann, Patrick Maeder
Summary: Nature has inspired scientists to develop new methods based on observations, with recent advances allowing insights into biological neural processes. Homeostatic plasticity, particularly synaptic scaling, has been identified as a mature and applicable theory to enhance learning capabilities of neural networks. Analyzing mutual information affected by synaptic scaling, the proposed approach outperforms previous regularization techniques in regression and classification tasks across various network topologies and data sets.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Multidisciplinary Sciences
Gangtao Xin, Pingyi Fan
Summary: With the advancement of intelligent vision algorithms and devices, image reprocessing and secondary propagation are becoming increasingly prevalent. Soft compression, a novel data-driven image coding algorithm, shows superior performance in contrast to traditional methods and holds promise for human-centric/data-centric intelligent systems. This paper presents a comprehensive analysis of soft compression and reveals the functional role of each component in the system.
SCIENTIFIC REPORTS
(2023)
Article
Computer Science, Information Systems
Tapas Bhadra, Sanghamitra Bandyopadhyay
Summary: The paper proposes a novel supervised feature selection approach based on dense subgraph discovery. The algorithm proceeds in two phases to select features with maximal average class relevance, minimal average pairwise redundancy, and good discriminating power. Experimental results show the proposed approach is competitive with conventional and state-of-the-art algorithms in supervised feature selection.
INFORMATION SCIENCES
(2021)
Article
Engineering, Electrical & Electronic
Shayan Hassanpour, Dirk Wuebben, Armin Dekorsy
Summary: This article introduces a generic iterative algorithm, M-FAVIB, for the multiterminal Joint Source-Channel Coding problem, and demonstrates the effectiveness of purely statistical design approaches obtained through the Information Bottleneck method.
IEEE TRANSACTIONS ON COMMUNICATIONS
(2021)
Article
Automation & Control Systems
Reza Zamani, Mohsen Parsa-Moghaddam, Mahmoud-Reza Haghifam
Summary: This article investigates the data transmission and communication issues in the transactive energy framework and proves the low sparsity of the prosumers dataset. It proposes a dynamic intelligent algorithm based on mutual information theorem and two data compression algorithms. Experimental results demonstrate that the proposed algorithms provide excellent performance while reducing information loss.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Article
Multidisciplinary Sciences
Qian Zeng, Jin Wang
Summary: This study focuses on refining the concept of Maxwell's demon to explore the limit of energy dissipation in open systems, uncovering a previously unexplored set of fluctuation theorems. These theorems reveal the existence of an intrinsic nonequilibrium state in the system, guided by nonnegative demon-induced dissipative information. The analysis suggests that the bounds of both work and heat in the system are tighter than previously predicted, and proposes a potential experimental test to verify these boundaries.
Article
Biochemical Research Methods
Snehalika Lall, Sumanta Ray, Sanghamitra Bandyopadhyay
Summary: The study introduces a method RgCop based on regularized copula for stable and predictive gene selection in large-scale single cell RNA sequencing data, improving clustering/classification performance and enhancing the robustness of the method.
PLOS COMPUTATIONAL BIOLOGY
(2021)
Article
Psychology, Biological
Arthur Prat-Carrabin, Michael Woodford
Summary: Humans weight different stimuli in averaging tasks, possibly indicating encoding bias. In a study, participants were asked to compare the averages of two series of numbers with varying prior distributions. It was found that participants encoded numbers with bias and noise, with more noise for infrequently occurring numbers.
NATURE HUMAN BEHAVIOUR
(2022)
Article
Computer Science, Artificial Intelligence
Abdelmalik Moujahid, Fadi Dornaika, Yassine Ruichek, Karim Hammoudi
NEURAL COMPUTING & APPLICATIONS
(2019)
Article
Computer Science, Information Systems
A. Moujahid, F. Dornaika
MULTIMEDIA TOOLS AND APPLICATIONS
(2019)
Article
Computer Science, Artificial Intelligence
Abdelmalik Moujahid, Fadi Dornaika
NEURAL COMPUTING & APPLICATIONS
(2020)
Article
Computer Science, Artificial Intelligence
Fadi Dornaika, Kunwei Wang, Ignacio Arganda-Carreras, Anne Elorza, Abdelmalik Moujahid
EXPERT SYSTEMS WITH APPLICATIONS
(2020)
Article
Automation & Control Systems
F. Dornaika, A. Moujahid, K. Wang, X. Feng
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2020)
Article
Computer Science, Artificial Intelligence
Abdelmalik Moujahid, Fadi Dornaika, Ignacio Arganda-Carreras, Jorge Reta
Summary: The study introduces a compact face monitoring system based on a compact face texture descriptor to effectively capture drowsy features. By utilizing a multi-scale pyramidal face representation and feature selection process, it shows promising results in drowsiness detection when compared to other methods.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Mathematics, Interdisciplinary Applications
A. Moujahid, F. Vadillo
Summary: Mathematical modeling is crucial for studying the impact of delay in neural systems and evaluating its effects on the signaling activity of coupled neurons. This study focuses on the energy perspective of delayed coupling in Hindmarsh-Rose burst neurons, examining the average energy consumption required to maintain cooperative behavior and quantifying the contribution of synapses to total energy consumption.
CHAOS SOLITONS & FRACTALS
(2022)
Article
Chemistry, Analytical
Pendar Alirezazadeh, Fadi Dornaika, Abdelmalik Moujahid
Summary: This article introduces the cross-domain discriminative margin loss (DML) to deal with the variability of negative pairs in the fashion domain. The experiments conducted on the Consumer-to-Shop Clothes Retrieval dataset confirm that the proposed loss function not only outperforms existing loss functions but also achieves the best performance.
Article
Computer Science, Artificial Intelligence
Fadi Dornaika, Abdelmalik Moujahid
Summary: The study introduces a graph-based semi-supervised method to find an appropriate graph representation of face images through the construction of multiple graphs. By combining geometric and deep features, it produces a higher-level representation of face images and shows superiority over other methods in experimental results.
Article
Computer Science, Artificial Intelligence
Pendar Alirezazadeh, Fadi Dornaika, Abdelmalik Moujahid
Summary: This paper proposes an end-to-end method for deep visual recognition in fashion style recognition and face analysis tasks, using Additive Cosine Margin Loss (ACML) as the novel loss function. Experimental results demonstrate that the proposed loss function outperforms existing methods on two fashion style recognition datasets and face verification datasets.
APPLIED SOFT COMPUTING
(2022)
Article
Environmental Sciences
Miriam Wahbi, Insaf El Bakali, Badia Ez-zahouani, Rida Azmi, Abdelmalik Moujahid, Mohammed Zouiten, Otmane Yazidi Alaoui, Hakim Boulaassal, Mustapha Maatouk, Omar El Kharki
Summary: Buildings in rural areas of Morocco present diverse shapes and sizes, posing challenges for accurate detection using optical satellite imagery and traditional image processing techniques. This study aims to detect and map settlements in the Souss-Massa region using deep learning algorithms with Sentinel-2 satellite images. The performance of the convolutional neural network architecture UNet was tested, followed by an evaluation of the impact of filters number on UNet's performance. The deep Residual UNet (ResUNet) was also implemented. Special metrics, including precision, recall, F1-score, and the ROC curve, were used to evaluate the quality of the tested models. The UNet with an increased number of filters achieved a precision of 87% and an F1-score of 54%, outperforming other algorithms such as UNet and ResUNet.
REMOTE SENSING APPLICATIONS-SOCIETY AND ENVIRONMENT
(2023)
Article
Multidisciplinary Sciences
Abdelmalik Moujahid, Fernando Vadillo
Summary: Ordinary differential equations (ODEs) are important for modeling and understanding various real systems. However, in the life sciences, time-delayed differential equations (DDEs) are often required for mathematical modeling. These models involve time delays that represent the manifestation of hidden processes, such as incubation periods or immune periods. Stochastic delay differential equations (SDDEs) provide a more realistic approximation to these models due to the uncertainties in biological systems. This study examines the predator-prey system using three types of time-delay models, revealing the significance of stochasticity and the magnitude of the delay in determining the system dynamics.
Article
Computer Science, Information Systems
Pendar Alirezazadeh, Fadi Dornaika, Abdelmalik Moujahid
Summary: This article discusses the use of different loss functions for classifying malignant and benign cancer in breast histopathological image datasets. The research results show that certain angular margin-based softmax loss functions can better extract distinguishing features, thus improving classification performance.
Article
Mathematics, Interdisciplinary Applications
Abdelmalik Moujahid, Fernando Vadillo
Summary: In many scientific fields, it is common to know the dynamics of a system and the main challenge lies in estimating the parameters that model the system behavior. This paper addresses this problem by building probabilistic stochastic differential models and using the data from these models to estimate the parameters through nonlinear regression analysis.
FRACTAL AND FRACTIONAL
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
F. Dornaika, J. Reta, I. Arganda-Carreras, A. Moujahid
2018 EIGHTH INTERNATIONAL CONFERENCE ON IMAGE PROCESSING THEORY, TOOLS AND APPLICATIONS (IPTA)
(2018)
Article
Biology
Theodoros Kyriazos, Mary Poga
Summary: This paper explores the innovative intersection of quantum mechanics and psychology, examining the potential impact of quantum principles on human emotions, cognition, and consciousness. By drawing parallels between quantum phenomena and psychological counterparts, a quantum-psychological model is proposed, reimagining the characteristics of emotional states, cognitive breakthroughs, interpersonal relationships, and the nature of consciousness. Computational models and simulations are used to explore the implications and applications of this interdisciplinary fusion, highlighting its potential benefits and inherent challenges. Approaching this emerging framework with both enthusiasm and skepticism is crucial, and rigorous empirical validation is necessary to fully realize its potential in research and therapeutic contexts.
Article
Biology
Abir U. Igamberdiev
Summary: Biological systems strive to maximize self-maintenance and adaptability by establishing stable non-equilibrium states that organize the fluxes of matter and energy and control metabolic processes. These states are realized in autopoietic structures that operate based on biological codes. The principle of thermodynamic buffering optimizes metabolic fluxes, and in developing systems, the principle transforms into increasing external work. Bauer's concept of the stable non-equilibrium state places thermodynamics within the framework of internal biological causality, providing a relational theory of biological thermodynamics.
Article
Biology
Oleg Gaidai, Vladimir Yakimov, Yuhao Niu, Zirui Liu
Summary: This study presents a new methodology for assessing pandemic risks in a national health system. The suggested approach addresses the highdimensionality and complex cross-correlations between regional observations, enabling accurate epidemiological risk forecasts for multi-regional biological and health systems.
Article
Biology
Amirreza Khalili Golmankhaneh, Suemeyye Tunc, Agnieszka Matylda Schlichtinger, Dachel Martinez Asanza, Alireza Khalili Golmankhaneh
Summary: This article introduces important concepts such as fractal calculus and fractal analysis, the calculation of squared residuals, and the determination of Aikaike's information criterion for fitting cancer-related data. The study also investigates the double-size cancer in the fractal temporal dimension with respect to various mathematical models.
