Frame Replication and Elimination for Reliability (FRER) in Time-Sensitive Networking (TSN) enhances fault tolerance by duplicating critical traffic across disjoint paths. However, always-on FRER configurations introduce persistent redundancy overhead, even under nominal network conditions. This paper proposes a predictive FRER activation framework that anticipates faults using a Key Performance Indicator (KPI)-driven bidirectional Long Short-Term Memory (BiLSTM) model. By continuously analyzing multivariate KPIs—such as latency, jitter, and retransmission rates—the model forecasts potential faults and proactively activates FRER. Redundancy is deactivated upon KPI recovery or after a defined minimum protection window, thereby reducing bandwidth usage without compromising reliability. The framework includes a Python-based simulation environment, a real-time visualization dashboard built with Streamlit, and a fully integrated runtime controller. The experimental results demonstrate substantial improvements in link utilization while preserving fault protection, highlighting the effectiveness of anticipatory redundancy strategies in industrial TSN environments.
Despite the potential the ability to identify granular materials creates for applications such as robotic cooking or earthmoving, granular material identification remains a challenging area, existing methods mostly relying on shaking the materials in closed containers. This work presents an interactive material identification framework that enables robots to identify a wide range of granular materials using only force-torque measurements. Unlike prior works, the proposed approach uses direct interaction with the materials. The approach is evaluated through experiments with a real-world dataset comprising 11 granular materials, which we also make publicly available. Results show that our method can identify a wide range of granular materials with near-perfect accuracy while relying solely on force measurements obtained from direct interaction. Further, our comprehensive data analysis and experiments show that a high-performancefeature space must combine features related to the force signal's time-domain dynamics and frequency spectrum. We account for this by proposing a combination of the raw signal and its high-frequency magnitude histogram as the suggested feature space representation. We show that the proposed feature space outperforms baselines by a significant margin. The code and data set are available at: https://irobotics.aalto.fi/identify_granular/.
Thanh Thi Nguyen, Campbell Wilson, Imad Khan
et al.
Forensic science plays a crucial role in legal investigations, and the use of advanced technologies, such as object detection based on machine learning methods, can enhance the efficiency and accuracy of forensic analysis. Human hands are unique and can leave distinct patterns, marks, or prints that can be utilized for forensic examinations. This paper compares various machine learning approaches to hand detection and presents the application results of employing the best-performing model to identify images of significant importance in forensic contexts. We fine-tune YOLOv8 and vision transformer-based object detection models on four hand image datasets, including the 11k hands dataset with our own bounding boxes annotated by a semi-automatic approach. Two YOLOv8 variants, i.e., YOLOv8 nano (YOLOv8n) and YOLOv8 extra-large (YOLOv8x), and two vision transformer variants, i.e., DEtection TRansformer (DETR) and Detection Transformers with Assignment (DETA), are employed for the experiments. Experimental results demonstrate that the YOLOv8 models outperform DETR and DETA on all datasets. The experiments also show that YOLOv8 approaches result in superior performance compared with existing hand detection methods, which were based on YOLOv3 and YOLOv4 models. Applications of our fine-tuned YOLOv8 models for identifying hand images (or frames in a video) with high forensic values produce excellent results, significantly reducing the time required by forensic experts. This implies that our approaches can be implemented effectively for real-world applications in forensics or related fields.
The transformer neural network has significantly out-shined all other neural network architectures as the engine behind large language models. We provide a theoretical analysis of the expressivity of the transformer architecture through the lens of topos theory. From this viewpoint, we show that many common neural network architectures, such as the convolutional, recurrent and graph convolutional networks, can be embedded in a pretopos of piecewise-linear functions, but that the transformer necessarily lives in its topos completion. In particular, this suggests that the two network families instantiate different fragments of logic: the former are first order, whereas transformers are higher-order reasoners. Furthermore, we draw parallels with architecture search and gradient descent, integrating our analysis in the framework of cybernetic agents.
Unmanned Aerial Vehicle (UAV) remote sensing is a commonly used technical means in modern science and technology, but currently, remote sensing images captured by UAVs need to be spliced to obtain more comprehensive information. However, current image stitching techniques generally have shortcomings such as a small number of extracted features, low matching accuracy, and poor stability. To address the above issues, this study proposes an improved remote sensing image mosaic model on the bias of the Scale Invariant Feature Transform (SIFT) algorithm. Firstly, in this study, aiming at the problem that traditional SIFT cannot meet the requirements of feature extraction and matching for unconventional remote sensing images and special texture images, normalized cross correlation (NCC) and Forstner operator are introduced to optimize it, namely, a SIFT-NCC model is constructed. On this basis, for remote sensing images with high resolution and a wide range, this study further proposes a remote sensing image automatic mosaic model that combines point features and line features. That is, a linear segment detector (LSD) is introduced to extract the line features of remote sensing images. The experimental verification results of the final SIFT-NCC-LSD show that the matching accuracy for remote sensing images with different characteristics can reach over 95 %. Therefore, SIFT-NCC-LSD has good applicability.
Sergey V. Skryl, Anastasiya A. Itskova, Kirill E. Ushakov
The article develops a functional model of unauthorized access (UA) protection mechanisms at information infrastructure objects (IIOs). It defines the content of protection measures, techniques used, and stages of their implementation. It substantiates the order of execution of individual functional components of the structural representation of the objective function "Protection of information of IIOs from UA". The sequence of implementation of these components is illustrated as a change in the states of the Markov process for constructing such a model. A table is provided of the correspondence between the list of procedures performed by an intruder in the process of implementing an UA threat to information of an IIOs and the procedures for protecting information, and it demonstrates the possibility of transition from the description of information protection measures from UA at IIOs in terms of functional modeling to the mathematical representation of the time characteristics of the functional components of the objective function of protection. The corresponding analytical expressions are provided for various options for representing the order of the functional components performed.
Background: Despite being significant contributors to the economy, informal businesses operate with limited resources. In South Africa, the informal sector is substantial, accounting for approximately 30% of total employment and around 6% of gross domestic product (GDP). These businesses often struggle to adopt and leverage technology constraining their capacity for growth and innovation, ultimately limiting their contribution to economic development and the alleviation of socio-economic challenges.
Objectives: The objective of this study was to investigate the factors that influence the barriers to adopting digital technologies in South Africa’s informal economy.
Method: This study adopted a qualitative research approach, using semi-structured interviews and purposive sampling to collect data from 14 informal business owners in Cape Town. Participants provided informed consent and thematic analysis was conducted using the Unified Theory of Acceptance and Use of Technology (UTAUT) framework.
Results: Findings revealed barriers including cash preference, load-shedding, crime and digital incompetency obstructing technology adoption. Despite these hurdles, the potential benefits of digital technology for informal businesses were underscored.
Conclusion: The study suggests untapped potential in enhancing technology adoption among informal businesses through targeted interventions. By addressing identified barriers, such initiatives could significantly bolster the informal economy’s impact on South Africa’s socio-economic landscape.
Contribution: This research contributes to understanding the complexities surrounding technology adoption in South Africa’s informal economy. It offers insights for policymakers, practitioners and stakeholders seeking to promote digital inclusion and economic empowerment within marginalised sectors.
Management information systems, Information theory
Results. The place and role of cybernetics methods for solving the global problem of directed evolution are considered. The author investigates the eventual phenomenon of the interaction of two coryphaei of Ukrainian science V.I. Vernadsky and V.M. Glushkov in the formation of these scientific fields and the synergistic effect of this interaction. A special attention is paid to the applied aspects of using the scientific theories when implementing complex topical processes at the state and global levels. The list of the applied fields of cybernetics with their implementation in the projects of the Institute of Cybernetics under the leadership of V.M. Glushkov is endless, and his scientific ideas are successfully brought to life today. Keywords: cognitive intelligent technology, transdisciplinary scientific research (TSR), systemology of TSR, scientific picture of the world, directed evolution, noospheric theory, ontological engineering, convergence of technologies, consolidated intelligence, collective consciousness, research design, eventual analysis.
Medical image segmentation is critical for disease diagnosis and treatment assessment. However, concerns regarding the reliability of segmentation regions persist among clinicians, mainly attributed to the absence of confidence assessment, robustness, and calibration to accuracy. To address this, we introduce DEviS, an easily implementable foundational model that seamlessly integrates into various medical image segmentation networks. DEviS not only enhances the calibration and robustness of baseline segmentation accuracy but also provides high-efficiency uncertainty estimation for reliable predictions. By leveraging subjective logic theory, we explicitly model probability and uncertainty for medical image segmentation. Here, the Dirichlet distribution parameterizes the distribution of probabilities for different classes of the segmentation results. To generate calibrated predictions and uncertainty, we develop a trainable calibrated uncertainty penalty. Furthermore, DEviS incorporates an uncertainty-aware filtering module, which designs the metric of uncertainty-calibrated error to filter out-of-distribution data. We conducted validation studies on publicly available datasets, including ISIC2018, KiTS2021, LiTS2017, and BraTS2019, to assess the accuracy and robustness of different backbone segmentation models enhanced by DEviS, as well as the efficiency and reliability of uncertainty estimation.
We extend known results on chordal graphs and distance-hereditary graphs to much larger graph classes by using only a common metric property of these graphs. Specifically, a graph is called $α_i$-metric ($i\in \mathcal{N}$) if it satisfies the following $α_i$-metric property for every vertices $u,w,v$ and $x$: if a shortest path between $u$ and $w$ and a shortest path between $x$ and $v$ share a terminal edge $vw$, then $d(u,x)\geq d(u,v) + d(v,x)-i$. Roughly, gluing together any two shortest paths along a common terminal edge may not necessarily result in a shortest path but yields a ``near-shortest'' path with defect at most $i$. It is known that $α_0$-metric graphs are exactly ptolemaic graphs, and that chordal graphs and distance-hereditary graphs are $α_i$-metric for $i=1$ and $i=2$, respectively. We show that an additive $O(i)$-approximation of the radius, of the diameter, and in fact of all vertex eccentricities of an $α_i$-metric graph can be computed in total linear time. Our strongest results are obtained for $α_1$-metric graphs, for which we prove that a central vertex can be computed in subquadratic time, and even better in linear time for so-called $(α_1,Δ)$-metric graphs (a superclass of chordal graphs and of plane triangulations with inner vertices of degree at least $7$). The latter answers a question raised in (Dragan, IPL, 2020). Our algorithms follow from new results on centers and metric intervals of $α_i$-metric graphs. In particular, we prove that the diameter of the center is at most $3i+2$ (at most $3$, if $i=1$). The latter partly answers a question raised in (Yushmanov & Chepoi, Mathematical Problems in Cybernetics, 1991).
Bashir Eseyin Abdullahi, Emeka Ogbuju, Taiwo Abiodun
et al.
Facial affective computing has gained popularity and become a progressive research area, as it plays a key role in human-computer interaction. However, many researchers lack the right technique to carry out a reliable facial affective computing effectively. To address this issue, we presented a review of the state-of-the-art artificial intelligence techniques that are being used for facial affective computing. Three research questions were answered by studying and analysing related papers collected from some well-established scientific databases based on some exclusion and inclusion criteria. The result presented the common artificial intelligence approaches for face detection, face recognition and emotion detection. The paper finds out that the haar-cascade algorithm has outperformed all the algorithms that have been used for face detection, the Convolutional Neural Network (CNN) based algorithms have performed best in face recognition, and the neural network algorithm with multiple layers has the best performance in emotion detection. A limitation of this research is the access to some research papers, as some documents require a high subscription cost.
Practice implication: The paper provides a comprehensive and unbiased analysis of existing literature, identifying knowledge gaps and future research direction and supports evidence-based decision-making. We considered articles and conference papers from well-established databases. The method presents a novel scope for facial affective computing and provides decision support for researchers when selecting plans for facial affective computing.
Special aspects of education, Electronic computers. Computer science
Large multi-agent systems are often driven by locally defined agent interactions, which is referred to as self-organization. Our primary objective is to determine when the propagation of such local interactions will reach a specific agent of interest. Although conventional approaches that reconstruct all agent states can be used, they may entail unnecessary computational costs. In this paper, we investigate a CNN-LSTM model to forecast the state of a particular agent in a large self-organizing multi-agent system without the reconstruction. The proposed model comprises a CNN encoder to represent the system in a low-dimensional vector, a LSTM module to learn agent dynamics in the vector space, and a MLP decoder to predict the future state of an agent. As an example, we consider a forest fire model where we aim to predict when a particular tree agent will start burning. We compare the proposed model with reconstruction-based approaches such as CNN-LSTM and ConvLSTM. The proposed model exhibits similar or slightly worse AUC but significantly reduces computational costs such as activation than ConvLSTM. Moreover, it achieves higher AUC with less computation than the recontruction-based CNN-LSTM.
The theoretical analysis of multi-class classification has proved that the existing multi-class classification methods can train a classifier with high classification accuracy on the test set, when the instances are precise in the training and test sets with same distribution and enough instances can be collected in the training set. However, one limitation with multi-class classification has not been solved: how to improve the classification accuracy of multi-class classification problems when only imprecise observations are available. Hence, in this paper, we propose a novel framework to address a new realistic problem called multi-class classification with imprecise observations (MCIMO), where we need to train a classifier with fuzzy-feature observations. Firstly, we give the theoretical analysis of the MCIMO problem based on fuzzy Rademacher complexity. Then, two practical algorithms based on support vector machine and neural networks are constructed to solve the proposed new problem. Experiments on both synthetic and real-world datasets verify the rationality of our theoretical analysis and the efficacy of the proposed algorithms.
The paper considers a problem of classifying Sentinel-2 multispectral satellite images for environmental monitoring of the Baikal Natural Territory (BNT). The specificity of the BNT required the creation of a new set of 12 classes, which takes into account current problems. The set was formed in such a way that the areas corresponding to these classes completely covered the BNT. A training dataset was formed using a web interface based on Sentinel-2 satellite images. The classification of satellite images was carried out using Random Forest algorithms and the ResNet50 neural network. The accuracy of the calculations showed that the classification results can be used to solve actual problems of the Baikal natural territory, in particular, to analyze changes in the forestland, assess the impact of climate change on the landscape, analyze the dynamics of development activities, create farmland inventory, etc.
Liudmila I. Khoruzhy, Yuriy N. Katkov, Ekaterina A. Katkova
et al.
The development of cloud technologies enables companies to actively implement technologies for cost management and risk reduction in their financial and economic activities. The use of cloud-based models of risk management in the financial and economic activities of the enterprise will help small and medium-sized companies in the agro-industrial sector in Russia to make structural and strategic changes, as well as discover new opportunities for business expansion. The purpose of the study is to develop models for cost management and reduction of risks in the financial and economic activities of companies based on the OLAP technology for application in Russian agro-industrial enterprises. The study employs a qualitative approach based on the case study methodology. The paper discloses and substantiates the authors’ conceptual model of a cost management system that allows executives to make decisions proceeding from four types of cost prices. The distinguishing feature of the management system is the use of a digital twin, which makes it possible to manage risks at the early stages of decision-making. The application of OLAP systems improves the quality of analysis and visualization methods as part of the cost management system. In addition, the study provides practical insight into how the applied model will help small and medium-sized agro-industrial enterprises to develop different business vision strategies based on cost reduction, manage the level of risk at the early stages of decision-making, and analyze information from a geographically dispersed logistics chain of divisions (production facilities, warehouses, stores).
Isabelle A. Zaugg, Anushah Hossain, Brendan Molloy
Digitally-disadvantaged languages face multiple inequities in the digital sphere including gaps in digital support that obstruct access for speakers, poorly-designed digital tools that negatively affect the integrity of languages and writing systems, and unique vulnerabilities to surveillance harms for speaker communities. This term captures the acutely uneven digital playing field for speakers of the world’s 7000+ languages.
We present new models of optimization-based task and motion planning (TAMP) for robotic pick-and-place (P&P), which plan action sequences and motion trajectory with low computational costs. We improved an existing state-of-the-art TAMP model integrated with the collision avoidance, which is formulated as a mixed-integer linear programing (MILP) problem. To enable the MILP solver to search for solutions efficiently, we introduced two approaches leveraging features of collision avoidance in robotic P&P. The first approach reduces number of binary variables, which are related to the collision avoidance of delivery objects, by reformulating them as continuous variables with additional hard constraints. These hard constraints maintain consistency by conditionally propagating binary values, which are related to the carry action state and collision avoidance of robots, to the reformulated continuous variables. The second approach is more aware of the branch-and-bound method, which is the fundamental algorithm of modern MILP solvers. This approach guides the MILP solver to find integer solutions with shallower branching by adding a soft constraint, which softly restricts a robot's routes around delivery objects. We demonstrate the effectiveness of the proposed approaches with a modern MILP solver.
Few-shot speaker adaptation is a specific Text-to-Speech (TTS) system that aims to reproduce a novel speaker's voice with a few training data. While numerous attempts have been made to the few-shot speaker adaptation system, there is still a gap in terms of speaker similarity to the target speaker depending on the amount of data. To bridge the gap, we propose GC-TTS which achieves high-quality speaker adaptation with significantly improved speaker similarity. Specifically, we leverage two geometric constraints to learn discriminative speaker representations. Here, a TTS model is pre-trained for base speakers with a sufficient amount of data, and then fine-tuned for novel speakers on a few minutes of data with two geometric constraints. Two geometric constraints enable the model to extract discriminative speaker embeddings from limited data, which leads to the synthesis of intelligible speech. We discuss and verify the effectiveness of GC-TTS by comparing it with popular and essential methods. The experimental results demonstrate that GC-TTS generates high-quality speech from only a few minutes of training data, outperforming standard techniques in terms of speaker similarity to the target speaker.
Currently, an increasing number of model pruning methods are proposed to resolve the contradictions between the computer powers required by the deep learning models and the resource-constrained devices. However, most of the traditional rule-based network pruning methods can not reach a sufficient compression ratio with low accuracy loss and are time-consuming as well as laborious. In this paper, we propose Automatic Block-wise and Channel-wise Network Pruning (ABCP) to jointly search the block-wise and channel-wise pruning action with deep reinforcement learning. A joint sample algorithm is proposed to simultaneously generate the pruning choice of each residual block and the channel pruning ratio of each convolutional layer from the discrete and continuous search space respectively. The best pruning action taking both the accuracy and the complexity of the model into account is obtained finally. Compared with the traditional rule-based pruning method, this pipeline saves human labor and achieves a higher compression ratio with lower accuracy loss. Tested on the mobile robot detection dataset, the pruned YOLOv3 model saves 99.5% FLOPs, reduces 99.5% parameters, and achieves 37.3 times speed up with only 2.8% mAP loss. The results of the transfer task on the sim2real detection dataset also show that our pruned model has much better robustness performance.