Kliknij ten link, aby zobaczyć inne rodzaje publikacji na ten temat: Metric learning paradigm.
Artykuły w czasopismach na temat „Metric learning paradigm”
Utwórz poprawne odniesienie w stylach APA, MLA, Chicago, Harvard i wielu innych
Sprawdź 50 najlepszych artykułów w czasopismach naukowych na temat „Metric learning paradigm”.
Przycisk „Dodaj do bibliografii” jest dostępny obok każdej pracy w bibliografii. Użyj go – a my automatycznie utworzymy odniesienie bibliograficzne do wybranej pracy w stylu cytowania, którego potrzebujesz: APA, MLA, Harvard, Chicago, Vancouver itp.
Możesz również pobrać pełny tekst publikacji naukowej w formacie „.pdf” i przeczytać adnotację do pracy online, jeśli odpowiednie parametry są dostępne w metadanych.
Przeglądaj artykuły w czasopismach z różnych dziedzin i twórz odpowiednie bibliografie.
1
Brockmeier, Austin J., John S. Choi, Evan G. Kriminger, Joseph T. Francis i Jose C. Principe. "Neural Decoding with Kernel-Based Metric Learning". Neural Computation 26, nr 6 (czerwiec 2014): 1080–107. http://dx.doi.org/10.1162/neco_a_00591.
Pełny tekst źródłaStreszczenie:
In studies of the nervous system, the choice of metric for the neural responses is a pivotal assumption. For instance, a well-suited distance metric enables us to gauge the similarity of neural responses to various stimuli and assess the variability of responses to a repeated stimulus—exploratory steps in understanding how the stimuli are encoded neurally. Here we introduce an approach where the metric is tuned for a particular neural decoding task. Neural spike train metrics have been used to quantify the information content carried by the timing of action potentials. While a number of metrics for individual neurons exist, a method to optimally combine single-neuron metrics into multineuron, or population-based, metrics is lacking. We pose the problem of optimizing multineuron metrics and other metrics using centered alignment, a kernel-based dependence measure. The approach is demonstrated on invasively recorded neural data consisting of both spike trains and local field potentials. The experimental paradigm consists of decoding the location of tactile stimulation on the forepaws of anesthetized rats. We show that the optimized metrics highlight the distinguishing dimensions of the neural response, significantly increase the decoding accuracy, and improve nonlinear dimensionality reduction methods for exploratory neural analysis.
2
Saha, Soumadeep, Utpal Garain, Arijit Ukil, Arpan Pal i Sundeep Khandelwal. "MedTric : A clinically applicable metric for evaluation of multi-label computational diagnostic systems". PLOS ONE 18, nr 8 (10.08.2023): e0283895. http://dx.doi.org/10.1371/journal.pone.0283895.
Pełny tekst źródłaStreszczenie:
When judging the quality of a computational system for a pathological screening task, several factors seem to be important, like sensitivity, specificity, accuracy, etc. With machine learning based approaches showing promise in the multi-label paradigm, they are being widely adopted to diagnostics and digital therapeutics. Metrics are usually borrowed from machine learning literature, and the current consensus is to report results on a diverse set of metrics. It is infeasible to compare efficacy of computational systems which have been evaluated on different sets of metrics. From a diagnostic utility standpoint, the current metrics themselves are far from perfect, often biased by prevalence of negative samples or other statistical factors and importantly, they are designed to evaluate general purpose machine learning tasks. In this paper we outline the various parameters that are important in constructing a clinical metric aligned with diagnostic practice, and demonstrate their incompatibility with existing metrics. We propose a new metric, MedTric that takes into account several factors that are of clinical importance. MedTric is built from the ground up keeping in mind the unique context of computational diagnostics and the principle of risk minimization, penalizing missed diagnosis more harshly than over-diagnosis. MedTric is a unified metric for medical or pathological screening system evaluation. We compare this metric against other widely used metrics and demonstrate how our system outperforms them in key areas of medical relevance.
3
Gong, Xiuwen, Dong Yuan i Wei Bao. "Online Metric Learning for Multi-Label Classification". Proceedings of the AAAI Conference on Artificial Intelligence 34, nr 04 (3.04.2020): 4012–19. http://dx.doi.org/10.1609/aaai.v34i04.5818.
Pełny tekst źródłaStreszczenie:
Existing research into online multi-label classification, such as online sequential multi-label extreme learning machine (OSML-ELM) and stochastic gradient descent (SGD), has achieved promising performance. However, these works lack an analysis of loss function and do not consider label dependency. Accordingly, to fill the current research gap, we propose a novel online metric learning paradigm for multi-label classification. More specifically, we first project instances and labels into a lower dimension for comparison, then leverage the large margin principle to learn a metric with an efficient optimization algorithm. Moreover, we provide theoretical analysis on the upper bound of the cumulative loss for our method. Comprehensive experiments on a number of benchmark multi-label datasets validate our theoretical approach and illustrate that our proposed online metric learning (OML) algorithm outperforms state-of-the-art methods.
4
Qiu, Wei. "Based on Semi-Supervised Clustering with the Boost Similarity Metric Method for Face Retrieval". Applied Mechanics and Materials 543-547 (marzec 2014): 2720–23. http://dx.doi.org/10.4028/www.scientific.net/amm.543-547.2720.
Pełny tekst źródłaStreszczenie:
The focus of this paper is on Metric Learning, with particular interest in incorporating side information to make it semi-supervised. This study is primarily motivated by an application: face-image clustering. In the paper introduces metric learning and semi-supervised clustering, Boost the similarity metric learning method that adapt the underlying similarity metric used by the clustering algorithm. we propose a novel idea of learning with historical relevance feedback log data, and adopt a new paradigm called Boost the Similarity Metric Method for Face Retrieval, Experimental results demonstrate that the unified approach produces better clusters than both individual approaches as well as previously proposed semi-supervised clustering algorithms. This paper followed by the discussion of experiments on face-image clustering.
5
Xiao, Qiao, Khuan Lee, Siti Aisah Mokhtar, Iskasymar Ismail, Ahmad Luqman bin Md Pauzi, Qiuxia Zhang i Poh Ying Lim. "Deep Learning-Based ECG Arrhythmia Classification: A Systematic Review". Applied Sciences 13, nr 8 (14.04.2023): 4964. http://dx.doi.org/10.3390/app13084964.
Pełny tekst źródłaStreszczenie:
Deep learning (DL) has been introduced in automatic heart-abnormality classification using ECG signals, while its application in practical medical procedures is limited. A systematic review is performed from perspectives of the ECG database, preprocessing, DL methodology, evaluation paradigm, performance metric, and code availability to identify research trends, challenges, and opportunities for DL-based ECG arrhythmia classification. Specifically, 368 studies meeting the eligibility criteria are included. A total of 223 (61%) studies use MIT-BIH Arrhythmia Database to design DL models. A total of 138 (38%) studies considered removing noise or artifacts in ECG signals, and 102 (28%) studies performed data augmentation to extend the minority arrhythmia categories. Convolutional neural networks are the dominant models (58.7%, 216) used in the reviewed studies while growing studies have integrated multiple DL structures in recent years. A total of 319 (86.7%) and 38 (10.3%) studies explicitly mention their evaluation paradigms, i.e., intra- and inter-patient paradigms, respectively, where notable performance degradation is observed in the inter-patient paradigm. Compared to the overall accuracy, the average F1 score, sensitivity, and precision are significantly lower in the selected studies. To implement the DL-based ECG classification in real clinical scenarios, leveraging diverse ECG databases, designing advanced denoising and data augmentation techniques, integrating novel DL models, and deeper investigation in the inter-patient paradigm could be future research opportunities.
6
Niu, Gang, Bo Dai, Makoto Yamada i Masashi Sugiyama. "Information-Theoretic Semi-Supervised Metric Learning via Entropy Regularization". Neural Computation 26, nr 8 (sierpień 2014): 1717–62. http://dx.doi.org/10.1162/neco_a_00614.
Pełny tekst źródłaStreszczenie:
We propose a general information-theoretic approach to semi-supervised metric learning called SERAPH (SEmi-supervised metRic leArning Paradigm with Hypersparsity) that does not rely on the manifold assumption. Given the probability parameterized by a Mahalanobis distance, we maximize its entropy on labeled data and minimize its entropy on unlabeled data following entropy regularization. For metric learning, entropy regularization improves manifold regularization by considering the dissimilarity information of unlabeled data in the unsupervised part, and hence it allows the supervised and unsupervised parts to be integrated in a natural and meaningful way. Moreover, we regularize SERAPH by trace-norm regularization to encourage low-dimensional projections associated with the distance metric. The nonconvex optimization problem of SERAPH could be solved efficiently and stably by either a gradient projection algorithm or an EM-like iterative algorithm whose M-step is convex. Experiments demonstrate that SERAPH compares favorably with many well-known metric learning methods, and the learned Mahalanobis distance possesses high discriminability even under noisy environments.
7
Wilde, Henry, Vincent Knight i Jonathan Gillard. "Evolutionary dataset optimisation: learning algorithm quality through evolution". Applied Intelligence 50, nr 4 (27.12.2019): 1172–91. http://dx.doi.org/10.1007/s10489-019-01592-4.
Pełny tekst źródłaStreszczenie:
AbstractIn this paper we propose a novel method for learning how algorithms perform. Classically, algorithms are compared on a finite number of existing (or newly simulated) benchmark datasets based on some fixed metrics. The algorithm(s) with the smallest value of this metric are chosen to be the ‘best performing’. We offer a new approach to flip this paradigm. We instead aim to gain a richer picture of the performance of an algorithm by generating artificial data through genetic evolution, the purpose of which is to create populations of datasets for which a particular algorithm performs well on a given metric. These datasets can be studied so as to learn what attributes lead to a particular progression of a given algorithm. Following a detailed description of the algorithm as well as a brief description of an open source implementation, a case study in clustering is presented. This case study demonstrates the performance and nuances of the method which we call Evolutionary Dataset Optimisation. In this study, a number of known properties about preferable datasets for the clustering algorithms known as k-means and DBSCAN are realised in the generated datasets.
8
Zhukov, Alexey, Jenny Benois-Pineau i Romain Giot. "Evaluation of Explanation Methods of AI - CNNs in Image Classification Tasks with Reference-based and No-reference Metrics". Advances in Artificial Intelligence and Machine Learning 03, nr 01 (2023): 620–46. http://dx.doi.org/10.54364/aaiml.2023.1143.
Pełny tekst źródłaStreszczenie:
The most popular methods in AI-machine learning paradigm are mainly black boxes. This is why explanation of AI decisions is of emergency. Although dedicated explanation tools have been massively developed, the evaluation of their quality remains an open research question. In this paper, we generalize the methodologies of evaluation of post-hoc explainers of CNNs’ decisions in visual classification tasks with reference and no-reference based metrics. We apply them on our previously developed explainers (FEM1 , MLFEM), and popular Grad-CAM. The reference-based metrics are Pearson correlation coefficient and Similarity computed between the explanation map and its ground truth represented by a Gaze Fixation Density Map obtained with a psycho-visual experiment. As a no-reference metric, we use stability metric, proposed by Alvarez-Melis and Jaakkola. We study its behaviour, consensus with reference-based metrics and show that in case of several kinds of degradation on input images, this metric is in agreement with reference-based ones. Therefore, it can be used for evaluation of the quality of explainers when the ground truth is not available.
9
Pinto, Danna, Anat Prior i Elana Zion Golumbic. "Assessing the Sensitivity of EEG-Based Frequency-Tagging as a Metric for Statistical Learning". Neurobiology of Language 3, nr 2 (2022): 214–34. http://dx.doi.org/10.1162/nol_a_00061.
Pełny tekst źródłaStreszczenie:
Abstract Statistical learning (SL) is hypothesized to play an important role in language development. However, the measures typically used to assess SL, particularly at the level of individual participants, are largely indirect and have low sensitivity. Recently, a neural metric based on frequency-tagging has been proposed as an alternative measure for studying SL. We tested the sensitivity of frequency-tagging measures for studying SL in individual participants in an artificial language paradigm, using non-invasive electroencephalograph (EEG) recordings of neural activity in humans. Importantly, we used carefully constructed controls to address potential acoustic confounds of the frequency-tagging approach, and compared the sensitivity of EEG-based metrics to both explicit and implicit behavioral tests of SL. Group-level results confirm that frequency-tagging can provide a robust indication of SL for an artificial language, above and beyond potential acoustic confounds. However, this metric had very low sensitivity at the level of individual participants, with significant effects found only in 30% of participants. Comparison of the neural metric to previously established behavioral measures for assessing SL showed a significant yet weak correspondence with performance on an implicit task, which was above-chance in 70% of participants, but no correspondence with the more common explicit 2-alternative forced-choice task, where performance did not exceed chance-level. Given the proposed ubiquitous nature of SL, our results highlight some of the operational and methodological challenges of obtaining robust metrics for assessing SL, as well as the potential confounds that should be taken into account when using the frequency-tagging approach in EEG studies.
10
Gomoluch, Paweł, Dalal Alrajeh i Alessandra Russo. "Learning Classical Planning Strategies with Policy Gradient". Proceedings of the International Conference on Automated Planning and Scheduling 29 (25.05.2021): 637–45. http://dx.doi.org/10.1609/icaps.v29i1.3531.
Pełny tekst źródłaStreszczenie:
A common paradigm in classical planning is heuristic forward search. Forward search planners often rely on simple best-first search which remains fixed throughout the search process. In this paper, we introduce a novel search framework capable of alternating between several forward search approaches while solving a particular planning problem. Selection of the approach is performed using a trainable stochastic policy, mapping the state of the search to a probability distribution over the approaches. This enables using policy gradient to learn search strategies tailored to a specific distributions of planning problems and a selected performance metric, e.g. the IPC score. We instantiate the framework by constructing a policy space consisting of five search approaches and a two-dimensional representation of the planner’s state. Then, we train the system on randomly generated problems from five IPC domains using three different performance metrics. Our experimental results show that the learner is able to discover domain-specific search strategies, improving the planner’s performance relative to the baselines of plain bestfirst search and a uniform policy.
11
Dou, Jason Xiaotian, Lei Luo i Raymond Mingrui Yang. "An Optimal Transport Approach to Deep Metric Learning (Student Abstract)". Proceedings of the AAAI Conference on Artificial Intelligence 36, nr 11 (28.06.2022): 12935–36. http://dx.doi.org/10.1609/aaai.v36i11.21604.
Pełny tekst źródłaStreszczenie:
Capturing visual similarity among images is the core of many computer vision and pattern recognition tasks. This problem can be formulated in such a paradigm called metric learning. Most research in the area has been mainly focusing on improving the loss functions and similarity measures. However, due to the ignoring of geometric structure, existing methods often lead to sub-optimal results. Thus, several recent research methods took advantage of Wasserstein distance between batches of samples to characterize the spacial geometry. Although these approaches can achieve enhanced performance, the aggregation over batches definitely hinders Wasserstein distance's superior measure capability and leads to high computational complexity. To address this limitation, we propose a novel Deep Wasserstein Metric Learning framework, which employs Wasserstein distance to precisely capture the relationship among various images under ranking-based loss functions such as contrastive loss and triplet loss. Our method directly computes the distance between images, considering the geometry at a finer granularity than batch level. Furthermore, we introduce a new efficient algorithm using Sinkhorn approximation and Wasserstein measure coreset. The experimental results demonstrate the improvements of our framework over various baselines in different applications and benchmark datasets.
12
Wang, Yabin, Zhiheng Ma, Zhiwu Huang, Yaowei Wang, Zhou Su i Xiaopeng Hong. "Isolation and Impartial Aggregation: A Paradigm of Incremental Learning without Interference". Proceedings of the AAAI Conference on Artificial Intelligence 37, nr 8 (26.06.2023): 10209–17. http://dx.doi.org/10.1609/aaai.v37i8.26216.
Pełny tekst źródłaStreszczenie:
This paper focuses on the prevalent stage interference and stage performance imbalance of incremental learning. To avoid obvious stage learning bottlenecks, we propose a new incremental learning framework, which leverages a series of stage-isolated classifiers to perform the learning task at each stage, without interference from others. To be concrete, to aggregate multiple stage classifiers as a uniform one impartially, we first introduce a temperature-controlled energy metric for indicating the confidence score levels of the stage classifiers. We then propose an anchor-based energy self-normalization strategy to ensure the stage classifiers work at the same energy level. Finally, we design a voting-based inference augmentation strategy for robust inference. The proposed method is rehearsal-free and can work for almost all incremental learning scenarios. We evaluate the proposed method on four large datasets. Extensive results demonstrate the superiority of the proposed method in setting up new state-of-the-art overall performance. Code is available at https://github.com/iamwangyabin/ESN.
13
Ge, Ce, Jingyu Wang, Qi Qi, Haifeng Sun, Tong Xu i Jianxin Liao. "Semi-transductive Learning for Generalized Zero-Shot Sketch-Based Image Retrieval". Proceedings of the AAAI Conference on Artificial Intelligence 37, nr 6 (26.06.2023): 7678–86. http://dx.doi.org/10.1609/aaai.v37i6.25931.
Pełny tekst źródłaStreszczenie:
Sketch-based image retrieval (SBIR) is an attractive research area where freehand sketches are used as queries to retrieve relevant images. Existing solutions have advanced the task to the challenging zero-shot setting (ZS-SBIR), where the trained models are tested on new classes without seen data. However, they are prone to overfitting under a realistic scenario when the test data includes both seen and unseen classes. In this paper, we study generalized ZS-SBIR (GZS-SBIR) and propose a novel semi-transductive learning paradigm. Transductive learning is performed on the image modality to explore the potential data distribution within unseen classes, and zero-shot learning is performed on the sketch modality sharing the learned knowledge through a semi-heterogeneous architecture. A hybrid metric learning strategy is proposed to establish semantics-aware ranking property and calibrate the joint embedding space. Extensive experiments are conducted on two large-scale benchmarks and four evaluation metrics. The results show that our method is superior over the state-of-the-art competitors in the challenging GZS-SBIR task.
14
De Santis, Enrico, Alessio Martino i Antonello Rizzi. "On component-wise dissimilarity measures and metric properties in pattern recognition". PeerJ Computer Science 8 (10.10.2022): e1106. http://dx.doi.org/10.7717/peerj-cs.1106.
Pełny tekst źródłaStreszczenie:
In many real-world applications concerning pattern recognition techniques, it is of utmost importance the automatic learning of the most appropriate dissimilarity measure to be used in object comparison. Real-world objects are often complex entities and need a specific representation grounded on a composition of different heterogeneous features, leading to a non-metric starting space where Machine Learning algorithms operate. However, in the so-called unconventional spaces a family of dissimilarity measures can be still exploited, that is, the set of component-wise dissimilarity measures, in which each component is treated with a specific sub-dissimilarity that depends on the nature of the data at hand. These dissimilarities are likely to be non-Euclidean, hence the underlying dissimilarity matrix is not isometrically embeddable in a standard Euclidean space because it may not be structurally rich enough. On the other hand, in many metric learning problems, a component-wise dissimilarity measure can be defined as a weighted linear convex combination and weights can be suitably learned. This article, after introducing some hints on the relation between distances and the metric learning paradigm, provides a discussion along with some experiments on how weights, intended as mathematical operators, interact with the Euclidean behavior of dissimilarity matrices.
15
Jaiswal, Mimansa, i Emily Mower Provost. "Privacy Enhanced Multimodal Neural Representations for Emotion Recognition". Proceedings of the AAAI Conference on Artificial Intelligence 34, nr 05 (3.04.2020): 7985–93. http://dx.doi.org/10.1609/aaai.v34i05.6307.
Pełny tekst źródłaStreszczenie:
Many mobile applications and virtual conversational agents now aim to recognize and adapt to emotions. To enable this, data are transmitted from users' devices and stored on central servers. Yet, these data contain sensitive information that could be used by mobile applications without user's consent or, maliciously, by an eavesdropping adversary. In this work, we show how multimodal representations trained for a primary task, here emotion recognition, can unintentionally leak demographic information, which could override a selected opt-out option by the user. We analyze how this leakage differs in representations obtained from textual, acoustic, and multimodal data. We use an adversarial learning paradigm to unlearn the private information present in a representation and investigate the effect of varying the strength of the adversarial component on the primary task and on the privacy metric, defined here as the inability of an attacker to predict specific demographic information. We evaluate this paradigm on multiple datasets and show that we can improve the privacy metric while not significantly impacting the performance on the primary task. To the best of our knowledge, this is the first work to analyze how the privacy metric differs across modalities and how multiple privacy concerns can be tackled while still maintaining performance on emotion recognition.
16
Yuan, Fei, Longtu Zhang, Huang Bojun i Yaobo Liang. "Simpson's Bias in NLP Training". Proceedings of the AAAI Conference on Artificial Intelligence 35, nr 16 (18.05.2021): 14276–83. http://dx.doi.org/10.1609/aaai.v35i16.17679.
Pełny tekst źródłaStreszczenie:
In most machine learning tasks, we evaluate a model M on a given data population S by measuring a population-level metric F(S;M). Examples of such evaluation metric F include precision/recall for (binary) recognition, the F1 score for multi-class classification, and the BLEU metric for language generation. On the other hand, the model M is trained by optimizing a sample-level loss G(S_t; M) at each learning step t, where S_t is a subset of S (a.k.a. the mini-batch). Popular choices of G include cross-entropy loss, the Dice loss, and sentence-level BLEU scores. A fundamental assumption behind this paradigm is that the mean value of the sample-level loss G, if averaged over all possible samples, should effectively represent the population-level metric F of the task, such as, that E[ G(S_t; M) ] ~ F(S; M). In this paper, we systematically investigate the above assumption in several NLP tasks. We show, both theoretically and experimentally, that some popular designs of the sample-level loss G may be inconsistent with the true population-level metric F of the task, so that models trained to optimize the former can be substantially sub-optimal to the latter, a phenomenon we call it, Simpson's bias, due to its deep connections with the classic paradox known as Simpson's reversal paradox in statistics and social sciences.
17
Khan, Koffka, i Wayne Goodridge. "Comparative study of One-Shot Learning in Dynamic Adaptive Streaming over HTTP : A Taxonomy-Based Analysis". International Journal of Advanced Networking and Applications 15, nr 01 (2023): 5822–30. http://dx.doi.org/10.35444/ijana.2023.15112.
Pełny tekst źródłaStreszczenie:
Dynamic Adaptive Streaming over HTTP (DASH) has revolutionized multimedia content delivery, enabling efficient video streaming over the internet. One-shot learning, a machine learning paradigm that allows recognition of new classes or objects with minimal training examples, holds promise for enhancing DASH systems. In this comparative study, we present a taxonomy-based analysis of one-shot learning techniques in the context of DASH, examining four taxonomies to provide a comprehensive understanding of their applications, evaluation metrics, and datasets. The first taxonomy focuses on categorizing one-shot learning techniques, including siamese networks, metric learning approaches, prototype-based methods, and generative models. This taxonomy reveals the diversity of techniques employed to tackle one-shot learning challenges in DASH environments. The second taxonomy explores the applications of one-shot learning in DASH. It highlights areas such as video quality prediction, buffer management, content adaptation, and bandwidth estimation, shedding light on how one-shot learning can optimize streaming decisions based on limited or single examples. The third taxonomy addresses evaluation metrics for one-shot learning in DASH. It encompasses accuracy-based metrics, generalization metrics, latency-related metrics, and robustness metrics, providing insights into the performance and effectiveness of one-shot learning approaches under various evaluation criteria. The fourth taxonomy delves into dataset characteristics for one-shot learning in DASH. It categorizes datasets into synthetic datasets, real-world datasets, transfer learning datasets, and unconstrained datasets, enabling researchers to select appropriate data sources and evaluate one-shot learning techniques in diverse streaming scenarios. By conducting this taxonomy-based analysis, our study provides researchers and practitioners with a structured framework for understanding and comparing different aspects of one-shot learning in DASH. It highlights the strengths, weaknesses, and potential applications of various techniques, offers guidance on evaluation metrics, and showcases dataset characteristics for benchmarking and future research. Ultimately, this comparative study aims to foster progress in one-shot learning for DASH by facilitating knowledge exchange, inspiring new research directions, and promoting the development of efficient and adaptive multimedia streaming systems over HTTP.
18
WOLFMAN, STEVEN A., i DANIEL S. WELD. "Combining linear programming and satisfiability solving for resource planning". Knowledge Engineering Review 16, nr 1 (marzec 2001): 85–99. http://dx.doi.org/10.1017/s0269888901000017.
Pełny tekst źródłaStreszczenie:
Compilation to Boolean satisfiability has become a powerful paradigm for solving artificial intelligence problems. However, domains that require metric reasoning cannot be compiled efficiently to satisfiability even if they would otherwise benefit from compilation. We address this problem by combining techniques from the artificial intelligence and operations research communities. In particular, we introduce the LCNF (Linear Conjunctive Normal Form) representation that combines propositional logic with metric constraints. We present LPSAT (Linear Programming plus SATisfiability), an engine that solves LCNF problems by interleaving calls to an incremental Simplex algorithm with systematic satisfaction methods. We explore several techniques for enhancing LPSAT's efficiency and expressive power by adjusting the interaction between the satisfiability and linear programming components of LPSAT. Next, we describe a compiler that converts metric resource planning problems into LCNF for processing by LPSAT. Finally, the experimental section of the paper explores several optimisations to LPSAT, including learning from constraint failure and randomised cutoffs.
19
Lin, Jianman, Jiantao Lin, Yuefang Gao, Zhijing Yang i Tianshui Chen. "Webly Supervised Fine-Grained Image Recognition with Graph Representation and Metric Learning". Electronics 11, nr 24 (11.12.2022): 4127. http://dx.doi.org/10.3390/electronics11244127.
Pełny tekst źródłaStreszczenie:
The aim of webly supervised fine-grained image recognition (FGIR) is to distinguish sub-ordinate categories based on data retrieved from the Internet, which can significantly mitigate the dependence of deep learning on manually annotated labels. Most current fine-grained image recognition algorithms use a large-scale data-driven deep learning paradigm, which relies heavily on manually annotated labels. However, there is a large amount of weakly labeled free data on the Internet. To utilize fine-grained web data effectively, this paper proposes a Graph Representation and Metric Learning (GRML) framework to learn discriminative and effective holistic–local features by graph representation for web fine-grained images and to handle noisy labels simultaneously, thus effectively using webly supervised data for training. Specifically, we first design an attention-focused module to locate the most discriminative region with different spatial aspects and sizes. Next, a structured instance graph is constructed to correlate holistic and local features to model the holistic–local information interaction, while a graph prototype that contains both holistic and local information for each category is introduced to learn category-level graph representation to assist in processing the noisy labels. Finally, a graph matching module is further employed to explore the holistic–local information interaction through intra-graph node information propagation as well as to evaluate the similarity score between each instance graph and its corresponding category-level graph prototype through inter-graph node information propagation. Extensive experiments were conducted on three webly supervised FGIR benchmark datasets, Web-Bird, Web-Aircraft and Web-Car, with classification accuracy of 76.62%, 85.79% and 82.99%, respectively. In comparison with Peer-learning, the classification accuracies of the three datasets separately improved 2.47%, 4.72% and 1.59%.
20
Samann, Fady Esmat Fathel, Adnan Mohsin Abdulazeez i Shavan Askar. "Fog Computing Based on Machine Learning: A Review". International Journal of Interactive Mobile Technologies (iJIM) 15, nr 12 (18.06.2021): 21. http://dx.doi.org/10.3991/ijim.v15i12.21313.
Pełny tekst źródłaStreszczenie:
<p>Internet of Things (IoT) systems usually produce massive amounts of data, while the number of devices connected to the internet might reach billions by now. Sending all this data over the internet will overhead the cloud and consume bandwidth. Fog computing's (FC) promising technology can solve the issue of computing and networking bottlenecks in large-scale IoT applications. This technology complements the cloud computing by providing processing power and storage to the edge of the network. However, it still suffers from performance and security issues. Thus, machine learning (ML) attracts attention for enabling FC to settle its issues. Lately, there has been a growing trend in utilizing ML to improve FC applications, like resource management, security, lessen latency and power usage. Also, intelligent FC was studied to address issues in industry 4.0, bioinformatics, blockchain and vehicular communication system. Due to the ML vital role in the FC paradigm, this work will shed light on recent studies utilized ML in a FC environment. Background knowledge about ML and FC also presented. This paper categorized the surveyed studies into three groups according to the aim of ML implementation. These studies were thoroughly reviewed and compared using sum-up tables. The results showed that not all studies used the same performance metric except those worked on security issues. In conclusion, the simulations of proposed ML models are not sufficient due to the heterogeneous nature of the FC paradigm.</p>
21
Elfakharany, Ahmed, i Zool Hilmi Ismail. "End-to-End Deep Reinforcement Learning for Decentralized Task Allocation and Navigation for a Multi-Robot System". Applied Sciences 11, nr 7 (24.03.2021): 2895. http://dx.doi.org/10.3390/app11072895.
Pełny tekst źródłaStreszczenie:
In this paper, we present a novel deep reinforcement learning (DRL) based method that is used to perform multi-robot task allocation (MRTA) and navigation in an end-to-end fashion. The policy operates in a decentralized manner mapping raw sensor measurements to the robot’s steering commands without the need to construct a map of the environment. We also present a new metric called the Task Allocation Index (TAI), which measures the performance of a method that performs MRTA and navigation from end-to-end in performing MRTA. The policy was trained on a simulated gazebo environment. The centralized learning and decentralized execution paradigm was used for training the policy. The policy was evaluated quantitatively and visually. The simulation results showed the effectiveness of the proposed method deployed on multiple Turtlebot3 robots.
22
Sotiropoulos, Dionisios N., Efthimios Alepis, Katerina Kabassi, Maria K. Virvou, George A. Tsihrintzis i Evangelos Sakkopoulos. "Artificial Immune System-Based Learning Style Stereotypes". International Journal on Artificial Intelligence Tools 28, nr 04 (czerwiec 2019): 1940008. http://dx.doi.org/10.1142/s0218213019400086.
Pełny tekst źródłaStreszczenie:
This paper addresses the problem of extracting fundamental learning style stereotypes through the exploitation of the biologically-inspired pattern recognition paradigm of Artificial Immune Systems (AIS). We present an unsupervised computational mechanism which exhibits the ability to reveal the inherent group structure of learning patterns that pervade a given set of educational profiles. We rely on the construction of an Artificial Immune Network (AIN) of learning style exemplars by proposing a correlation-based distance metric. This choice is actually imposed by the categoric nature of the underlying data. Our work utilizes an original dataset which was derived during the conduction of an extended empirical study involving students of the Hellenic Open University. The educational profiles of the students were built by collecting their answers on a thoroughly designed questionnaire taking into account a wide range of personal characteristics and skills. The efficiency of the proposed approach was assessed in terms of cluster compactness. Specifically, we measured the average correlation deviation of the students’ education profiles from the corresponding artificial memory antibodies that represent the acquired learning style stereotypes. Finally, the unsupervised learning procedure adopted in this paper was tested against a correlation-based version of the k-means algorithm indicating a significant improvement in performance for the AIS-based clustering approach.
23
Mwata-Velu, Tat’y, Juan Gabriel Avina-Cervantes, Jose Ruiz-Pinales, Tomas Alberto Garcia-Calva, Erick-Alejandro González-Barbosa, Juan B. Hurtado-Ramos i José-Joel González-Barbosa. "Improving Motor Imagery EEG Classification Based on Channel Selection Using a Deep Learning Architecture". Mathematics 10, nr 13 (1.07.2022): 2302. http://dx.doi.org/10.3390/math10132302.
Pełny tekst źródłaStreszczenie:
Recently, motor imagery EEG signals have been widely applied in Brain–Computer Interfaces (BCI). These signals are typically observed in the first motor cortex of the brain, resulting from the imagination of body limb movements. For non-invasive BCI systems, it is not apparent how to locate the electrodes, optimizing the accuracy for a given task. This study proposes a comparative analysis of channel signals exploiting the Deep Learning (DL) technique and a public dataset to locate the most discriminant channels. EEG channels are usually selected based on the function and nomenclature of electrode location from international standards. Instead, the most suitable configuration for a given paradigm must be determined by analyzing the proper selection of the channels. Therefore, an EEGNet network was implemented to classify signals from different channel location using the accuracy metric. Achieved results were then contrasted with results from the state-of-the-art. As a result, the proposed method improved BCI classification accuracy.
24
Liu, Pingping, Guixia Gou, Xue Shan, Dan Tao i Qiuzhan Zhou. "Global Optimal Structured Embedding Learning for Remote Sensing Image Retrieval". Sensors 20, nr 1 (4.01.2020): 291. http://dx.doi.org/10.3390/s20010291.
Pełny tekst źródłaStreszczenie:
A rich line of works focus on designing elegant loss functions under the deep metric learning (DML) paradigm to learn a discriminative embedding space for remote sensing image retrieval (RSIR). Essentially, such embedding space could efficiently distinguish deep feature descriptors. So far, most existing losses used in RSIR are based on triplets, which have disadvantages of local optimization, slow convergence and insufficient use of similarity structure in a mini-batch. In this paper, we present a novel DML method named as global optimal structured loss to deal with the limitation of triplet loss. To be specific, we use a softmax function rather than a hinge function in our novel loss to realize global optimization. In addition, we present a novel optimal structured loss, which globally learn an efficient deep embedding space with mined informative sample pairs to force the positive pairs within a limitation and push the negative ones far away from a given boundary. We have conducted extensive experiments on four public remote sensing datasets and the results show that the proposed global optimal structured loss with pairs mining scheme achieves the state-of-the-art performance compared with the baselines.
25
Li, Hui, Jinhao Liu i Dian Wang. "A Fast Instance Segmentation Technique for Log End Faces Based on Metric Learning". Forests 14, nr 4 (13.04.2023): 795. http://dx.doi.org/10.3390/f14040795.
Pełny tekst źródłaStreszczenie:
The diameter of the logs on a vehicle is a critical part of the logistics and transportation of logs. However, the manual size-checking method is inefficient and affects the efficiency of log transportation. The example segmentation methods can generate masks for each log end face, which helps automate the check gauge of logs and improve efficiency. The example segmentation model uses rectangle detection to identify each end face and then traverses the rectangular boxes for mask extraction. The traversal of rectangular boxes increases the time consumption of the model and lacks separate handling of the overlapping areas between rectangular boxes, which causes a decline in mask extraction accuracy. To address the above problems, we propose a fast instance segmentation method for further improving the efficiency and accuracy of log-checking diameter. The method uses a convolutional neural network to extract the mask image, rectangular frame prediction image, and embed the vector image from the input image. The mask image is used to extract the log end face region, and the rectangular frame prediction image generates an enveloping rectangular frame for each log, which in turn divides the log end face region into instances. For the overlapping regions of rectangular boxes, a metric learning paradigm is used to increase the embedding vector distance between pixels located in different logs and decrease the embedding vector distance between pixels of the same log, and finally the mask pixels of the overlapping regions of rectangular boxes are instantiated according to the pixel embedding vectors. This method avoids repeated calls to the contour extraction algorithm for each rectangular box and enables fine delineation of pixels in the overlapping rectangular box region. To verify the efficiency of the proposed algorithm, the log working pile is photographed in different scenes using a smartphone to obtain the end face recognition database and divide the training set, validation set, and test set according to 8:1:1. Secondly, the proposed model is used to obtain log end face masks, and the log end face ruler diameter is determined by an edge-fitting algorithm combined with a ruler. Finally, the practicality of the algorithm is evaluated by calculating the check-rule diameter error, running speed, and the error of wood volume calculation under different national standards. The proposed method has 91.2% and 50.2 FPS of mask extraction accuracy and running speed, respectively, which are faster and more accurate than the mainstream instance segmentation model. The relative error of the proposed method is −4.62% for the check-rule diameter and −4.25%, −5.02%, −6.32%, and −5.73% for the wood volume measurement under the Chinese, Russian, American, and Japanese raw wood volume calculation standards, respectively. Among them, the error of the calculated timber volume according to our standard is the smallest, which indicates that the model in this paper is more suitable for application in domestic log production operations.
26
Motaung, William B., Kingsley A. Ogudo i Chabalala S. Chabalala. "Optimal Video Compression Parameter Tuning for Digital Video Broadcasting (DVB) using Deep Reinforcement Learning". International Conference on Intelligent and Innovative Computing Applications 2022 (31.12.2022): 270–76. http://dx.doi.org/10.59200/iconic.2022.030.
Pełny tekst źródłaStreszczenie:
DVB (digital video broadcasting) has undergone an enormous paradigm shift, especially through internet streaming that utilizes multiple channels (i.e., secured hypertext transfer protocols). However, due to the limitations of the current communication network infrastructure, video signals need to be compressed before transmission. Whereas most recent research has concentrated and focused on assessing video quality, little to no study has worked on improving the compression processes of digital video signals in lightweight DVB setups. This study provides a video compression strategy (DRL-VC) that employs deep reinforcement learning for learning the suitable parameters used in digital video signal compression. The problem is formulated as a multi-objective one, considering the structural similarity index metric (SSIM), the delay time, and the peak signal-to-noise ratio (PSNR). Based on the findings of the experiments, our proposed scheme increases bitrate savings while at a constant PSNR. Results also show that our scheme performs better than the benchmarked compression schemes. Finally, the root means square error values show a consistent rate across different video streams, indicating the validity of our proposed compression scheme.
27
Benvenuto, Giovana A., Marilaine Colnago, Maurício A. Dias, Rogério G. Negri, Erivaldo A. Silva i Wallace Casaca. "A Fully Unsupervised Deep Learning Framework for Non-Rigid Fundus Image Registration". Bioengineering 9, nr 8 (5.08.2022): 369. http://dx.doi.org/10.3390/bioengineering9080369.
Pełny tekst źródłaStreszczenie:
In ophthalmology, the registration problem consists of finding a geometric transformation that aligns a pair of images, supporting eye-care specialists who need to record and compare images of the same patient. Considering the registration methods for handling eye fundus images, the literature offers only a limited number of proposals based on deep learning (DL), whose implementations use the supervised learning paradigm to train a model. Additionally, ensuring high-quality registrations while still being flexible enough to tackle a broad range of fundus images is another drawback faced by most existing methods in the literature. Therefore, in this paper, we address the above-mentioned issues by introducing a new DL-based framework for eye fundus registration. Our methodology combines a U-shaped fully convolutional neural network with a spatial transformation learning scheme, where a reference-free similarity metric allows the registration without assuming any pre-annotated or artificially created data. Once trained, the model is able to accurately align pairs of images captured under several conditions, which include the presence of anatomical differences and low-quality photographs. Compared to other registration methods, our approach achieves better registration outcomes by just passing as input the desired pair of fundus images.
28
Sakakushev, Boris E., Blagoi I. Marinov, Penka P. Stefanova, Stefan St Kostianev i Evangelos K. Georgiou. "Striving for Better Medical Education: the Simulation Approach". Folia Medica 59, nr 2 (1.06.2017): 123–31. http://dx.doi.org/10.1515/folmed-2017-0039.
Pełny tekst źródłaStreszczenie:
AbstractMedical simulation is a rapidly expanding area within medical education due to advances in technology, significant reduction in training hours and increased procedural complexity. Simulation training aims to enhance patient safety through improved technical competency and eliminating human factors in a risk free environment. It is particularly applicable to a practical, procedure-orientated specialties.Simulation can be useful for novice trainees, experienced clinicians (e.g. for revalidation) and team building. It has become a cornerstone in the delivery of medical education, being a paradigm shift in how doctors are educated and trained. Simulation must take a proactive position in the development of metric-based simulation curriculum, adoption of proficiency benchmarking definitions, and should not depend on the simulation platforms used.Conversely, ingraining of poor practice may occur in the absence of adequate supervision, and equipment malfunction during the simulation can break the immersion and disrupt any learning that has occurred. Despite the presence of high technology, there is a substantial learning curve for both learners and facilitators.The technology of simulation continues to advance, offering devices capable of improved fidelity in virtual reality simulation, more sophisticated procedural practice and advanced patient simulators. Simulation-based training has also brought about paradigm shifts in the medical and surgical education arenas and ensured that the scope and impact of simulation will continue to broaden.
29
Kuang, Jiachen, Tangfei Tao, Qingqiang Wu, Chengcheng Han, Fan Wei, Shengchao Chen, Wenjie Zhou, Cong Yan i Guanghua Xu. "Domain-Adaptive Prototype-Recalibrated Network with Transductive Learning Paradigm for Intelligent Fault Diagnosis under Various Limited Data Conditions". Sensors 22, nr 17 (30.08.2022): 6535. http://dx.doi.org/10.3390/s22176535.
Pełny tekst źródłaStreszczenie:
In real industrial scenarios, intelligent fault diagnosis based on data-driven methods has been widely researched in the past decade. However, data scarcity is widespread in fault diagnosis tasks owning to the difficulties in collecting adequate data. As a result, there is an increasing demand for both researchers and engineers for fault identification with scarce data. To address this issue, an innovative domain-adaptive prototype-recalibrated network (DAPRN) based on a transductive learning paradigm and prototype recalibration strategy (PRS) is proposed, which has the potential to promote the generalization ability from the source domain to target domain in a few-shot fault diagnosis. Within this scheme, the DAPRN is composed of a feature extractor, a domain discriminator, and a label predictor. Concretely, the feature extractor is jointly optimized by the minimization of few-shot classification loss and the maximization of domain-discriminative loss. The cosine similarity-based label predictor, which is promoted by the PRS, is exploited to avoid the bias of naïve prototypes in the metric space and recognize the health conditions of machinery in the meta-testing process. The efficacy and advantage of DAPRN are validated by extensive experiments on bearing and gearbox datasets compared with seven popular and well-established few-shot fault diagnosis methods. In practical application, the proposed DAPRN is expected to solve more challenging few-shot fault diagnosis scenarios and facilitate practical fault identification problems in modern manufacturing.
30
Manzoor, Sumaira, Ye-Chan An, Gun-Gyo In, Yueyuan Zhang, Sangmin Kim i Tae-Yong Kuc. "SPT: Single Pedestrian Tracking Framework with Re-Identification-Based Learning Using the Siamese Model". Sensors 23, nr 10 (19.05.2023): 4906. http://dx.doi.org/10.3390/s23104906.
Pełny tekst źródłaStreszczenie:
Pedestrian tracking is a challenging task in the area of visual object tracking research and it is a vital component of various vision-based applications such as surveillance systems, human-following robots, and autonomous vehicles. In this paper, we proposed a single pedestrian tracking (SPT) framework for identifying each instance of a person across all video frames through a tracking-by-detection paradigm that combines deep learning and metric learning-based approaches. The SPT framework comprises three main modules: detection, re-identification, and tracking. Our contribution is a significant improvement in the results by designing two compact metric learning-based models using Siamese architecture in the pedestrian re-identification module and combining one of the most robust re-identification models for data associated with the pedestrian detector in the tracking module. We carried out several analyses to evaluate the performance of our SPT framework for single pedestrian tracking in the videos. The results of the re-identification module validate that our two proposed re-identification models surpass existing state-of-the-art models with increased accuracies of 79.2% and 83.9% on the large dataset and 92% and 96% on the small dataset. Moreover, the proposed SPT tracker, along with six state-of-the-art (SOTA) tracking models, has been tested on various indoor and outdoor video sequences. A qualitative analysis considering six major environmental factors verifies the effectiveness of our SPT tracker under illumination changes, appearance variations due to pose changes, changes in target position, and partial occlusions. In addition, quantitative analysis based on experimental results also demonstrates that our proposed SPT tracker outperforms the GOTURN, CSRT, KCF, and SiamFC trackers with a success rate of 79.7% while beating the DiamSiamRPN, SiamFC, CSRT, GOTURN, and SiamMask trackers with an average of 18 tracking frames per second.
31
Xu, Yanbing, Yanmei Zhang, Tingxuan Yue, Chengcheng Yu i Huan Li. "Graph-Based Domain Adaptation Few-Shot Learning for Hyperspectral Image Classification". Remote Sensing 15, nr 4 (18.02.2023): 1125. http://dx.doi.org/10.3390/rs15041125.
Pełny tekst źródłaStreszczenie:
Due to a lack of labeled samples, deep learning methods generally tend to have poor classification performance in practical applications. Few-shot learning (FSL), as an emerging learning paradigm, has been widely utilized in hyperspectral image (HSI) classification with limited labeled samples. However, the existing FSL methods generally ignore the domain shift problem in cross-domain scenes and rarely explore the associations between samples in the source and target domain. To tackle the above issues, a graph-based domain adaptation FSL (GDAFSL) method is proposed for HSI classification with limited training samples, which utilizes the graph method to guide the domain adaptation learning process in a uniformed framework. First, a novel deep residual hybrid attention network (DRHAN) is designed to extract discriminative embedded features efficiently for few-shot HSI classification. Then, a graph-based domain adaptation network (GDAN), which combines graph construction with domain adversarial strategy, is proposed to fully explore the domain correlation between source and target embedded features. By utilizing the fully explored domain correlations to guide the domain adaptation process, a domain invariant feature metric space is learned for few-shot HSI classification. Comprehensive experimental results conducted on three public HSI datasets demonstrate that GDAFSL is superior to the state-of-the-art with a small sample size.
32
Alshammari, Abdulaziz, i Rakan C. Chabaan. "Sppn-Rn101: Spatial Pyramid Pooling Network with Resnet101-Based Foreign Object Debris Detection in Airports". Mathematics 11, nr 4 (7.02.2023): 841. http://dx.doi.org/10.3390/math11040841.
Pełny tekst źródłaStreszczenie:
Over the past few years, aviation security has turned into a vital domain as foreign object debris (FOD) on the airport paved path possesses an enormous possible threat to airplanes at the time of takeoff and landing. Hence, FOD’s precise identification remains significant for assuring airplane flight security. The material features of FOD remain the very critical criteria for comprehending the destruction rate endured by an airplane. Nevertheless, the most frequent identification systems miss an efficient methodology for automated material identification. This study proffers a new FOD technique centered on transfer learning and also a mainstream deep convolutional neural network. For object detection (OD), this embraces the spatial pyramid pooling network with ResNet101 (SPPN-RN101), which assists in concatenating the local features upon disparate scales within a similar convolution layer with fewer position errors while identifying little objects. Additionally, Softmax with Adam Optimizer in CNN enhances the training speed with greater identification accuracy. This study presents FOD’s image dataset called FOD in Airports (FODA). In addition to the bounding boxes’ principal annotations for OD, FODA gives labeled environmental scenarios. Consequently, every annotation instance has been additionally classified into three light-level classes (bright, dim, and dark) and two weather classes (dry and wet). The proffered SPPN-ResNet101 paradigm is correlated to the former methodologies, and the simulation outcomes exhibit that the proffered study executes an AP medium of 0.55 for the COCO metric, 0.97 AP for the pascal metric, and 0.83 MAP of pascal metric.
33
Li, Shuyuan, Huabin Liu, Rui Qian, Yuxi Li, John See, Mengjuan Fei, Xiaoyuan Yu i Weiyao Lin. "TA2N: Two-Stage Action Alignment Network for Few-Shot Action Recognition". Proceedings of the AAAI Conference on Artificial Intelligence 36, nr 2 (28.06.2022): 1404–11. http://dx.doi.org/10.1609/aaai.v36i2.20029.
Pełny tekst źródłaStreszczenie:
Few-shot action recognition aims to recognize novel action classes (query) using just a few samples (support). The majority of current approaches follow the metric learning paradigm, which learns to compare the similarity between videos. Recently, it has been observed that directly measuring this similarity is not ideal since different action instances may show distinctive temporal distribution, resulting in severe misalignment issues across query and support videos. In this paper, we arrest this problem from two distinct aspects -- action duration misalignment and action evolution misalignment. We address them sequentially through a Two-stage Action Alignment Network (TA2N). The first stage locates the action by learning a temporal affine transform, which warps each video feature to its action duration while dismissing the action-irrelevant feature (e.g. background). Next, the second stage coordinates query feature to match the spatial-temporal action evolution of support by performing temporally rearrange and spatially offset prediction. Extensive experiments on benchmark datasets show the potential of the proposed method in achieving state-of-the-art performance for few-shot action recognition.
34
Alonso-Betanzos, Amparo, Verónica Bolón-Canedo, Guy R. Heyndrickx i Peter L. M. Kerkhof. "Exploring Guidelines for Classification of Major Heart Failure Subtypes by Using Machine Learning". Clinical Medicine Insights: Cardiology 9s1 (styczeń 2015): CMC.S18746. http://dx.doi.org/10.4137/cmc.s18746.
Pełny tekst źródłaStreszczenie:
Background Heart failure (HF) manifests as at least two subtypes. The current paradigm distinguishes the two by using both the metric ejection fraction (EF) and a constraint for end-diastolic volume. About half of all HF patients exhibit preserved EF. In contrast, the classical type of HF shows a reduced EF. Common practice sets the cut-off point often at or near EF = 50%, thus defining a linear divider. However, a rationale for this safe choice is lacking, while the assumption regarding applicability of strict linearity has not been justified. Additionally, some studies opt for eliminating patients from consideration for HF if 40 < EF < 50% (gray zone). Thus, there is a need for documented classification guidelines, solving gray zone ambiguity and formulating crisp delineation of transitions between phenotypes. Methods Machine learning (ML) models are applied to classify HF subtypes within the ventricular volume domain, rather than by the single use of EF. Various ML models, both unsupervised and supervised, are employed to establish a foundation for classification. Data regarding 48 HF patients are employed as training set for subsequent classification of Monte Carlo–generated surrogate HF patients ( n = 403). Next, we map consequences when EF cut-off differs from 50% (as proposed for women) and analyze HF candidates not covered by current rules. Results The training set yields best results for the Support Vector Machine method (test error 4.06%), covers the gray zone, and other clinically relevant HF candidates. End-systolic volume (ESV) emerges as a logical discriminator rather than EF as in the prevailing paradigm. Conclusions Selected ML models offer promise for classifying HF patients (including the gray zone), when driven by ventricular volume data. ML analysis indicates that ESV has a role in the development of guidelines to parse HF subtypes. The documented curvilinear relationship between EF and ESV suggests that the assumption concerning a linear EF divider may not be of general utility over the complete clinically relevant range.
35
Uzair, Muhammad, Mohsen Eskandari, Li Li i Jianguo Zhu. "Machine Learning Based Protection Scheme for Low Voltage AC Microgrids". Energies 15, nr 24 (12.12.2022): 9397. http://dx.doi.org/10.3390/en15249397.
Pełny tekst źródłaStreszczenie:
The microgrid (MG) is a popular concept to handle the high penetration of distributed energy resources, such as renewable and energy storage systems, into electric grids. However, the integration of inverter-interfaced distributed generation units (IIDGs) imposes control and protection challenges. Fault identification, classification and isolation are major concerns with IIDGs-based active MGs where IIDGs reveal arbitrary impedance and thus different fault characteristics. Moreover, bidirectional complex power flow creates extra difficulties for fault analysis. This makes the conventional methods inefficient, and a new paradigm in protection schemes is needed for IIDGs-dominated MGs. In this paper, a machine-learning (ML)-based protection technique is developed for IIDG-based AC MGs by extracting unique and novel features for detecting and classifying symmetrical and unsymmetrical faults. Different signals, namely, 400 samples, for wide variations in operating conditions of an MG are obtained through electromagnetic transient simulations in DIgSILENT PowerFactory. After retrieving and pre-processing the signals, 10 different feature extraction techniques, including new peaks metric and max factor, are applied to obtain 100 features. They are ranked using the Kruskal–Wallis H-Test to identify the best performing features, apart from estimating predictor importance for ensemble ML classification. The top 18 features are used as input to train 35 classification learners. Random Forest (RF) outperformed all other ML classifiers for fault detection and fault type classification with faulted phase identification. Compared to previous methods, the results show better performance of the proposed method.
36
Martinelli, M., C. J. A. P. Martins, S. Nesseris, D. Sapone, I. Tutusaus, A. Avgoustidis, S. Camera i in. "Euclid: Forecast constraints on the cosmic distance duality relation with complementary external probes". Astronomy & Astrophysics 644 (grudzień 2020): A80. http://dx.doi.org/10.1051/0004-6361/202039078.
Pełny tekst źródłaStreszczenie:
Context. In metric theories of gravity with photon number conservation, the luminosity and angular diameter distances are related via the Etherington relation, also known as the distance duality relation (DDR). A violation of this relation would rule out the standard cosmological paradigm and point to the presence of new physics. Aims. We quantify the ability of Euclid, in combination with contemporary surveys, to improve the current constraints on deviations from the DDR in the redshift range 0 < z < 1.6. Methods. We start with an analysis of the latest available data, improving previously reported constraints by a factor of 2.5. We then present a detailed analysis of simulated Euclid and external data products, using both standard parametric methods (relying on phenomenological descriptions of possible DDR violations) and a machine learning reconstruction using genetic algorithms. Results. We find that for parametric methods Euclid can (in combination with external probes) improve current constraints by approximately a factor of six, while for non-parametric methods Euclid can improve current constraints by a factor of three. Conclusions. Our results highlight the importance of surveys like Euclid in accurately testing the pillars of the current cosmological paradigm and constraining physics beyond the standard cosmological model.
37
Lyu, Yangxintong, Ionut Schiopu, Bruno Cornelis i Adrian Munteanu. "Framework for Vehicle Make and Model Recognition—A New Large-Scale Dataset and an Efficient Two-Branch–Two-Stage Deep Learning Architecture". Sensors 22, nr 21 (2.11.2022): 8439. http://dx.doi.org/10.3390/s22218439.
Pełny tekst źródłaStreszczenie:
In recent years, Vehicle Make and Model Recognition (VMMR) has attracted a lot of attention as it plays a crucial role in Intelligent Transportation Systems (ITS). Accurate and efficient VMMR systems are required in real-world applications including intelligent surveillance and autonomous driving. The paper introduces a new large-scale dataset and a novel deep learning paradigm for VMMR. A new large-scale dataset dubbed Diverse large-scale VMM (DVMM) is proposed collecting image-samples with the most popular vehicle brands operating in Europe. A novel VMMR framework is proposed which follows a two-branch architecture performing make and model recognition respectively. A two-stage training procedure and a novel decision module are proposed to process the make and model predictions and compute the final model prediction. In addition, a novel metric based on the true positive rate is proposed to compare classification confusion of the proposed 2B–2S and the baseline methods. A complex experimental validation is carried out, demonstrating the generality, diversity, and practicality of the proposed DVMM dataset. The experimental results show that the proposed framework provides 93.95% accuracy over the more diverse DVMM dataset and 95.85% accuracy over traditional VMMR datasets. The proposed two-branch approach outperforms the conventional one-branch approach for VMMR over small-, medium-, and large-scale datasets by providing lower vehicle model confusion and reduced inter-make ambiguity. The paper demonstrates the advantages of the proposed two-branch VMMR paradigm in terms of robustness and lower confusion relative to single-branch designs.
38
Amari, Shun-ichi, Hyeyoung Park i Tomoko Ozeki. "Singularities Affect Dynamics of Learning in Neuromanifolds". Neural Computation 18, nr 5 (maj 2006): 1007–65. http://dx.doi.org/10.1162/neco.2006.18.5.1007.
Pełny tekst źródłaStreszczenie:
The parameter spaces of hierarchical systems such as multilayer perceptrons include singularities due to the symmetry and degeneration of hidden units. A parameter space forms a geometrical manifold, called the neuromanifold in the case of neural networks. Such a model is identified with a statistical model, and a Riemannian metric is given by the Fisher information matrix. However, the matrix degenerates at singularities. Such a singular structure is ubiquitous not only in multilayer perceptrons but also in the gaussian mixture probability densities, ARMA time-series model, and many other cases. The standard statistical paradigm of the Cramér-Rao theorem does not hold, and the singularity gives rise to strange behaviors in parameter estimation, hypothesis testing, Bayesian inference, model selection, and in particular, the dynamics of learning from examples. Prevailing theories so far have not paid much attention to the problem caused by singularity, relying only on ordinary statistical theories developed for regular (nonsingular) models. Only recently have researchers remarked on the effects of singularity, and theories are now being developed. This article gives an overview of the phenomena caused by the singularities of statistical manifolds related to multilayer perceptrons and gaussian mixtures. We demonstrate our recent results on these problems. Simple toy models are also used to show explicit solutions. We explain that the maximum likelihood estimator is no longer subject to the gaussian distribution even asymptotically, because the Fisher information matrix degenerates, that the model selection criteria such as AIC, BIC, and MDL fail to hold in these models, that a smooth Bayesian prior becomes singular in such models, and that the trajectories of dynamics of learning are strongly affected by the singularity, causing plateaus or slow manifolds in the parameter space. The natural gradient method is shown to perform well because it takes the singular geometrical structure into account. The generalization error and the training error are studied in some examples.
39
Voulodimos, Athanasios, Eftychios Protopapadakis, Iason Katsamenis, Anastasios Doulamis i Nikolaos Doulamis. "A Few-Shot U-Net Deep Learning Model for COVID-19 Infected Area Segmentation in CT Images". Sensors 21, nr 6 (22.03.2021): 2215. http://dx.doi.org/10.3390/s21062215.
Pełny tekst źródłaStreszczenie:
Recent studies indicate that detecting radiographic patterns on CT chest scans can yield high sensitivity and specificity for COVID-19 identification. In this paper, we scrutinize the effectiveness of deep learning models for semantic segmentation of pneumonia-infected area segmentation in CT images for the detection of COVID-19. Traditional methods for CT scan segmentation exploit a supervised learning paradigm, so they (a) require large volumes of data for their training, and (b) assume fixed (static) network weights once the training procedure has been completed. Recently, to overcome these difficulties, few-shot learning (FSL) has been introduced as a general concept of network model training using a very small amount of samples. In this paper, we explore the efficacy of few-shot learning in U-Net architectures, allowing for a dynamic fine-tuning of the network weights as new few samples are being fed into the U-Net. Experimental results indicate improvement in the segmentation accuracy of identifying COVID-19 infected regions. In particular, using 4-fold cross-validation results of the different classifiers, we observed an improvement of 5.388 ± 3.046% for all test data regarding the IoU metric and a similar increment of 5.394 ± 3.015% for the F1 score. Moreover, the statistical significance of the improvement obtained using our proposed few-shot U-Net architecture compared with the traditional U-Net model was confirmed by applying the Kruskal-Wallis test (p-value = 0.026).
40
Anand, S. S., P. W. Hamilton, J. G. Hughes i D. A. Bell. "On Prognostic Models, Artificial Intelligence and Censored Observations". Methods of Information in Medicine 40, nr 01 (2001): 18–24. http://dx.doi.org/10.1055/s-0038-1634459.
Pełny tekst źródłaStreszczenie:
AbstractThe development of prognostic models for assisting medical practitioners with decision making is not a trivial task. Models need to possess a number of desirable characteristics and few, if any, current modelling approaches based on statistical or artificial intelligence can produce models that display all these characteristics. The inability of modelling techniques to provide truly useful models has led to interest in these models being purely academic in nature. This in turn has resulted in only a very small percentage of models that have been developed being deployed in practice. On the other hand, new modelling paradigms are being proposed continuously within the machine learning and statistical community and claims, often based on inadequate evaluation, being made on their superiority over traditional modelling methods. We believe that for new modelling approaches to deliver true net benefits over traditional techniques, an evaluation centric approach to their development is essential. In this paper we present such an evaluation centric approach to developing extensions to the basic k-nearest neighbour (k-NN) paradigm. We use standard statistical techniques to enhance the distance metric used and a framework based on evidence theory to obtain a prediction for the target example from the outcome of the retrieved exemplars. We refer to this new k-NN algorithm as Censored k-NN (Ck-NN). This reflects the enhancements made to k-NN that are aimed at providing a means for handling censored observations within k-NN.
41
YAN, YUHONG, i HAN LIANG. "LAZY LEARNER ON DECISION TREE FOR RANKING". International Journal on Artificial Intelligence Tools 17, nr 01 (luty 2008): 139–58. http://dx.doi.org/10.1142/s0218213008003819.
Pełny tekst źródłaStreszczenie:
This paper aims to improve probability-based ranking (e.g. AUC) under decision-tree paradigm. We observe the fact that probability-based ranking is to sort samples in terms of their class probabilities. Therefore, ranking is a relative evaluation metric among those samples. This motivates us to use a lazy learner to explicitly yield a set of unique class probabilities for a testing sample based on its similarities to the training samples within its neighborhood. We embed lazy learners at the leaves of a decision tree to give class probability assignments. This results in the first model, named Lazy Distance-based Tree (LDTree). Then we further improve this model by continuing to grow the tree for the second time, and call the resulting model Eager Distance-based Tree (EDTree). In addition to the benefits of lazy learning, EDTree also takes advantage of the finer resolution of a large tree structure. We compare our models with C4.5, C4.4 and their variants in AUC on a large suite of UCI sample sets. The improvement shows that our method follows a new path that leads to better ranking performance.
42
Rayala, Venkat, i Satyanarayan Reddy Kalli. "Big Data Clustering Using Improvised Fuzzy C-Means Clustering". Revue d'Intelligence Artificielle 34, nr 6 (31.12.2020): 701–8. http://dx.doi.org/10.18280/ria.340604.
Pełny tekst źródłaStreszczenie:
Clustering emerged as powerful mechanism to analyze the massive data generated by modern applications; the main aim of it is to categorize the data into clusters where objects are grouped into the particular category. However, there are various challenges while clustering the big data recently. Deep Learning has been powerful paradigm for big data analysis, this requires huge number of samples for training the model, which is time consuming and expensive. This can be avoided though fuzzy approach. In this research work, we design and develop an Improvised Fuzzy C-Means (IFCM)which comprises the encoder decoder Convolutional Neural Network (CNN) model and Fuzzy C-means (FCM) technique to enhance the clustering mechanism. Encoder decoder based CNN is used for learning feature and faster computation. In general, FCM, we introduce a function which measure the distance between the cluster center and instance which helps in achieving the better clustering and later we introduce Optimized Encoder Decoder (OED) CNN model for improvising the performance and for faster computation. Further in order to evaluate the proposed mechanism, three distinctive data types namely Modified National Institute of Standards and Technology (MNIST), fashion MNIST and United States Postal Service (USPS) are used, also evaluation is carried out by considering the performance metric like Accuracy, Adjusted Rand Index (ARI) and Normalized Mutual Information (NMI). Moreover, comparative analysis is carried out on each dataset and comparative analysis shows that IFCM outperforms the existing model.
43
Longo, Mathias, Matías Hirsch, Cristian Mateos i Alejandro Zunino. "Towards Integrating Mobile Devices into Dew Computing: A Model for Hour-Wise Prediction of Energy Availability". Information 10, nr 3 (26.02.2019): 86. http://dx.doi.org/10.3390/info10030086.
Pełny tekst źródłaStreszczenie:
With self-provisioning of resources as premise, dew computing aims at providing computing services by minimizing the dependency over existing internetwork back-haul. Mobile devices have a huge potential to contribute to this emerging paradigm, not only due to their proximity to the end user, ever growing computing/storage features and pervasiveness, but also due to their capability to render services for several hours, even days, without being plugged to the electricity grid. Nonetheless, misusing the energy of their batteries can discourage owners to offer devices as resource providers in dew computing environments. Arguably, having accurate estimations of remaining battery would help to take better advantage of a device’s computing capabilities. In this paper, we propose a model to estimate mobile devices battery availability by inspecting traces of real mobile device owner’s activity and relevant device state variables. The model includes a feature extraction approach to obtain representative features/variables, and a prediction approach, based on regression models and machine learning classifiers. On average, the accuracy of our approach, measured with the mean squared error metric, overpasses the one obtained by a related work. Prediction experiments at five hours ahead are performed over activity logs of 23 mobile users across several months.
44
Tamm, Markus-Oliver, Yar Muhammad i Naveed Muhammad. "Classification of Vowels from Imagined Speech with Convolutional Neural Networks". Computers 9, nr 2 (1.06.2020): 46. http://dx.doi.org/10.3390/computers9020046.
Pełny tekst źródłaStreszczenie:
Imagined speech is a relatively new electroencephalography (EEG) neuro-paradigm, which has seen little use in Brain-Computer Interface (BCI) applications. Imagined speech can be used to allow physically impaired patients to communicate and to use smart devices by imagining desired commands and then detecting and executing those commands in a smart device. The goal of this research is to verify previous classification attempts made and then design a new, more efficient neural network that is noticeably less complex (fewer number of layers) that still achieves a comparable classification accuracy. The classifiers are designed to distinguish between EEG signal patterns corresponding to imagined speech of different vowels and words. This research uses a dataset that consists of 15 subjects imagining saying the five main vowels (a, e, i, o, u) and six different words. Two previous studies on imagined speech classifications are verified as those studies used the same dataset used here. The replicated results are compared. The main goal of this study is to take the proposed convolutional neural network (CNN) model from one of the replicated studies and make it much more simpler and less complex, while attempting to retain a similar accuracy. The pre-processing of data is described and a new CNN classifier with three different transfer learning methods is described and used to classify EEG signals. Classification accuracy is used as the performance metric. The new proposed CNN, which uses half as many layers and less complex pre-processing methods, achieved a considerably lower accuracy, but still managed to outperform the initial model proposed by the authors of the dataset by a considerable margin. It is recommended that further studies investigating classifying imagined speech should use more data and more powerful machine learning techniques. Transfer learning proved beneficial and should be used to improve the effectiveness of neural networks.
45
Dave, Chitrak Vimalbhai. "An Efficient Framework for Cost and Effort Estimation of Scrum Projects". International Journal for Research in Applied Science and Engineering Technology 9, nr 11 (30.11.2021): 1478–87. http://dx.doi.org/10.22214/ijraset.2021.39030.
Pełny tekst źródłaStreszczenie:
Abstract: Software Process Models from its inception instill standardization and creates a generic culture of developing software for various IT industries. A great paradigm shift has been observed in terms of embracing Agile Development methodology as a viable development methodology in cross key business units. There is a buffet of agile methodologies comes under the umbrella of ASD, out of which Scrum got the highest popularity and acceptability index. Agile based software development is the need of immediate environment. There is an increasing demand for significant changes to software systems to meet ever-changing user requirements and specifications. As Agile is volatile, so effort estimation is challenging and still striving for perfection to decide size, effort, cost, duration and schedule of projects with minimum error. This cause sensitizes potential researchers all across the globe to start working on addressing the issue of inaccurate predication of efforts. The gap between estimated and actual effort is because of limited or no inclusion of various estimation factors like people and project related factors, inappropriate use of size metric and cost drivers, ignorance of testing effort, team member’s inability to understand user story size and complexity etc. This paper attempts to bridge the gap of estimated and actual effort by the use of soft computing techniques thus taking the research to advance frontier area in terms ofestimation. Keywords: Cost Estimation, Effort Estimation, Scrum, Machine Learning, Agile Software Development
46
Galindo-Noreña, Steven, David Cárdenas-Peña i Álvaro Orozco-Gutierrez. "Multiple Kernel Stein Spatial Patterns for the Multiclass Discrimination of Motor Imagery Tasks". Applied Sciences 10, nr 23 (2.12.2020): 8628. http://dx.doi.org/10.3390/app10238628.
Pełny tekst źródłaStreszczenie:
Brain–computer interface (BCI) systems communicate the human brain and computers by converting electrical activity into commands to use external devices. Such kind of system has become an alternative for interaction with the environment for people suffering from motor disabilities through the motor imagery (MI) paradigm. Despite being the most widespread, electroencephalography (EEG)-based MI systems are highly sensitive to noise and artifacts. Further, spatially close brain activity sources and variability among subjects hampers the system performance. This work proposes a methodology for the classification of EEG signals, termed Multiple Kernel Stein Spatial Patterns (MKSSP) dealing with noise, raveled brain activity, and subject variability issues. Firstly, a bank of bandpass filters decomposes brain activity into spectrally independent multichannel signals. Then, Multi-Kernel Stein Spatial Patterns (MKSSP) maps each signal into low-dimensional covariance matrices preserving the nonlinear channel relationships. The Stein kernel provides a parameterized similarity metric for covariance matrices that belong to a Riemannian manifold. Lastly, the multiple kernel learning assembles the similarities from each spectral decomposition into a single kernel matrix that feeds the classifier. Experimental evaluations in the well-known four-class MI dataset 2a BCI competition IV proves that the methodology significantly improves state-of-the-art approaches. Further, the proposal is interpretable in terms of data distribution, spectral relevance, and spatial patterns. Such interpretability demonstrates that MKSSP encodes features from different spectral bands into a single representation improving the discrimination of mental tasks.
47
Feng, Jialiang, i Jie Gong. "AoI-Aware Optimization of Service Caching-Assisted Offloading and Resource Allocation in Edge Cellular Networks". Sensors 23, nr 6 (21.03.2023): 3306. http://dx.doi.org/10.3390/s23063306.
Pełny tekst źródłaStreszczenie:
The rapid development of the Internet of Things (IoT) has led to computational offloading at the edge; this is a promising paradigm for achieving intelligence everywhere. As offloading can lead to more traffic in cellular networks, cache technology is used to alleviate the channel burden. For example, a deep neural network (DNN)-based inference task requires a computation service that involves running libraries and parameters. Thus, caching the service package is necessary for repeatedly running DNN-based inference tasks. On the other hand, as the DNN parameters are usually trained in distribution, IoT devices need to fetch up-to-date parameters for inference task execution. In this work, we consider the joint optimization of computation offloading, service caching, and the AoI metric. We formulate a problem to minimize the weighted sum of the average completion delay, energy consumption, and allocated bandwidth. Then, we propose the AoI-aware service caching-assisted offloading framework (ASCO) to solve it, which consists of the method of Lagrange multipliers with the KKT condition-based offloading module (LMKO), the Lyapunov optimization-based learning and update control module (LLUC), and the Kuhn–Munkres (KM) algorithm-based channel-division fetching module (KCDF). The simulation results demonstrate that our ASCO framework achieves superior performance in regard to time overhead, energy consumption, and allocated bandwidth. It is verified that our ASCO framework not only benefits the individual task but also the global bandwidth allocation.
48
Powers, David. "Unsupervised Learning of Linguistic Structure". International Journal of Corpus Linguistics 2, nr 1 (1.01.1997): 91–131. http://dx.doi.org/10.1075/ijcl.2.1.06pow.
Pełny tekst źródłaStreszczenie:
Computational Linguistics and Natural Language have long been targets for Machine Learning, and a variety of learning paradigms and techniques have been employed with varying degrees of success. In this paper, we review approaches which have adopted an unsupervised learning paradigm, explore the assumptions which underlie the techniques used, and develop an approach to empirical evaluation. We concentrate on a statistical framework based on N-grams, although we seek to maintain neurolinguistic plausibility. The model we adopt places putative linguistic units in focus and associates them with a characteristic vector of statistics derived from occurrence frequency. These vectors are treated as defining a hyperspace, within which we demonstrate a technique for examining the empirical utility of the various metrics and normalization, visualization, and clustering techniques proposed in the literature. We conclude with an evaluation of the relative utility of a large array of different metrics and processing techniques in relation to our defined performance criteria.
49
Azam, Abu Bakr, Yu Qing Chang, Matthew Leong Tze Ker, Denise Goh, Jeffrey Chun Tatt Lim, Mai Chan Lau, Benedict Tan, Lihui Huang, Joe Yeong i Yiyu Cai. "818 Using deep learning approaches with mIF images to enhance T cell identification for tumor -automation of infiltrating lymphocytes (TILs) scoring on H&E images". Journal for ImmunoTherapy of Cancer 9, Suppl 2 (listopad 2021): A855—A856. http://dx.doi.org/10.1136/jitc-2021-sitc2021.818.
Pełny tekst źródłaStreszczenie:
BackgroundExamining Hematoxylin & Eosin (H&E) images using brightfield microscopes is the gold standard of pathological diagnosis as it is an inexpensive method and provides basic information of tumors and other nuclei. Complementary to H&E-stained images, Immunohistochemical (IHC) images are crucial in identifying tumor subtypes and efficacy of treatment response. Other newer technologies such as Multiplex Immunofluorescence (mIF) in particular, identifies cells such as tumor infiltrating lymphocytes (TILs) which can be augmented via immunotherapy, an evolving form of cancer treatment. Immunotherapy helps in the manipulation of the host immune response and overcome limitations like the PD-1 (Programmed Cell Death-1) receptor induced restrictions on TIL production. If the same biopsy specimen is used for inspection, the higher order features in H&E images can be used to obtain information usually found in mIF images using Convolutional Neural Networks (CNNs), widely used in object detection and image segmentation tasks.MethodsAs shown in (figure 1), firstly, a novel optical flow-based image registration paradigm is prepared to co-register H&E and mIF image pairs, aided by adaptive color thresholding and automated color clustering. Secondly, generative adversarial networks (GANs) are adapted to predict TIL (CD3, CD45) regions. For this purpose, a unique dataset is ideated and used in which a given single channel mIF image, e.g., a CD3 channel mIF image is superimposed on the corresponding H&E image. Primarily, the Pix2Pix GAN model is used to predict CD3 and/or CD45 regions.ResultsThe intensity-based image registration workflow is fast and fully compatible with the given dataset, with an increase in evaluation metric scores after alignment (table 1). Furthermore, this study would be the first implementation of optical flow as the registration algorithm for pathological images. Next, the use of the special dataset not only reduces penalization during the training of the Pix2Pix model, but also helped in gaining repeatable results with high scores in metrics like structural similarity index measure and peak-signal to noise ratio, with minimal effects on location accuracy (table 2 and table 3).ConclusionsThis multi-modal pathological image transformation study could potentially reduce dependence on mIF and IHC images for TILs scoring, reducing the amount of tissue and cost needed for examination, as its information is derived directly from inexpensive H&E images automatically – ultimately develop into a pathologist-assisted tool for TILs scoring. This would be highly beneficial in facilities where resources are relatively limited.Ethics ApprovalThe Agency of Science, Technology and Research, Singapore, provided approval for the use of control tissue materials in this study IRB: 2020 112Abstract 818 Figure 1Proposed workflowAbstract 818 Table 1Image registration metricsAbstract 818 Table 2CD3 negative regions examplesAbstract 818 Table 3CD3 positive regions examples
50
Samtani, Sagar, Yidong Chai i Hsinchun Chen. "Linking Exploits from the Dark Web to Known Vulnerabilities for Proactive Cyber Threat Intelligence: An Attention-Based Deep Structured Semantic Model". MIS Quarterly 46, nr 2 (24.05.2022): 911–46. http://dx.doi.org/10.25300/misq/2022/15392.
Pełny tekst źródłaStreszczenie:
Black hat hackers use malicious exploits to circumvent security controls and take advantage of system vulnerabilities worldwide, costing the global economy over $450 billion annually. While many organizations are increasingly turning to cyber threat intelligence (CTI) to help prioritize their vulnerabilities, extant CTI processes are often criticized as being reactive to known exploits. One promising data source that can help develop proactive CTI is the vast and ever-evolving Dark Web. In this study, we adopted the computational design science paradigm to design a novel deep learning (DL)- based exploit-vulnerability attention deep structured semantic model (EVA-DSSM) that includes bidirectional processing and attention mechanisms to automatically link exploits from the Dark Web to vulnerabilities. We also devised a novel device vulnerability severity metric (DVSM) that incorporates the exploit post date and vulnerability severity to help cybersecurity professionals with their device prioritization and risk management efforts. We rigorously evaluated the EVA-DSSM against state-of-theart non-DL and DL-based methods for short text matching on 52,590 exploit-vulnerability linkages across four testbeds: web application, remote, local, and denial of service. Results of these evaluations indicate that the proposed EVA-DSSM achieves precision at 1 scores 20% - 41% higher than non-DL approaches and 4% - 10% higher than DL-based approaches. We demonstrated the EVA-DSSM’s and DVSM’s practical utility with two CTI case studies: openly accessible systems in the top eight U.S. hospitals and over 20,000 Supervisory Control and Data Acquisition (SCADA) systems worldwide. A complementary user evaluation of the case study results indicated that 45 cybersecurity professionals found the EVADSSM and DVSM results more useful for exploit-vulnerability linking and risk prioritization activities than those produced by prevailing approaches. Given the rising cost of cyberattacks, the EVA-DSSM and DVSM have important implications for analysts in security operations centers, incident response teams, and cybersecurity vendors.