Добірка наукової літератури з теми "SELF-ATTENTION MECHANISM"
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся зі списками актуальних статей, книг, дисертацій, тез та інших наукових джерел на тему "SELF-ATTENTION MECHANISM".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Статті в журналах з теми "SELF-ATTENTION MECHANISM"
1
Yang, Kehua, Yaodong Wang, Wei Zhang, Jiqing Yao, and Yuquan Le. "Keyphrase Generation Based on Self-Attention Mechanism." Computers, Materials & Continua 61, no. 2 (2019): 569–81. http://dx.doi.org/10.32604/cmc.2019.05952.
Повний текст джерела
2
Liu, Siqi, Jiangshu Wei, Gang Liu, and Bei Zhou. "Image classification model based on large kernel attention mechanism and relative position self-attention mechanism." PeerJ Computer Science 9 (April 21, 2023): e1344. http://dx.doi.org/10.7717/peerj-cs.1344.
Повний текст джерелаАнотація:
The Transformer has achieved great success in many computer vision tasks. With the in-depth exploration of it, researchers have found that Transformers can better obtain long-range features than convolutional neural networks (CNN). However, there will be a deterioration of local feature details when the Transformer extracts local features. Although CNN is adept at capturing the local feature details, it cannot easily obtain the global representation of features. In order to solve the above problems effectively, this paper proposes a hybrid model consisting of CNN and Transformer inspired by Visual Attention Net (VAN) and CoAtNet. This model optimizes its shortcomings in the difficulty of capturing the global representation of features by introducing Large Kernel Attention (LKA) in CNN while using the Transformer blocks with relative position self-attention variant to alleviate the problem of detail deterioration in local features of the Transformer. Our model effectively combines the advantages of the above two structures to obtain the details of local features more accurately and capture the relationship between features far apart more efficiently on a large receptive field. Our experiments show that in the image classification task without additional training data, the proposed model in this paper can achieve excellent results on the cifar10 dataset, the cifar100 dataset, and the birds400 dataset (a public dataset on the Kaggle platform) with fewer model parameters. Among them, SE_LKACAT achieved a Top-1 accuracy of 98.01% on the cifar10 dataset with only 7.5M parameters.
3
Zhu, Hu, Ze Wang, Yu Shi, Yingying Hua, Guoxia Xu, and Lizhen Deng. "Multimodal Fusion Method Based on Self-Attention Mechanism." Wireless Communications and Mobile Computing 2020 (September 23, 2020): 1–8. http://dx.doi.org/10.1155/2020/8843186.
Повний текст джерелаАнотація:
Multimodal fusion is one of the popular research directions of multimodal research, and it is also an emerging research field of artificial intelligence. Multimodal fusion is aimed at taking advantage of the complementarity of heterogeneous data and providing reliable classification for the model. Multimodal data fusion is to transform data from multiple single-mode representations to a compact multimodal representation. In previous multimodal data fusion studies, most of the research in this field used multimodal representations of tensors. As the input is converted into a tensor, the dimensions and computational complexity increase exponentially. In this paper, we propose a low-rank tensor multimodal fusion method with an attention mechanism, which improves efficiency and reduces computational complexity. We evaluate our model through three multimodal fusion tasks, which are based on a public data set: CMU-MOSI, IEMOCAP, and POM. Our model achieves a good performance while flexibly capturing the global and local connections. Compared with other multimodal fusions represented by tensors, experiments show that our model can achieve better results steadily under a series of attention mechanisms.
4
Cao, Fude, Chunguang Zheng, Limin Huang, Aihua Wang, Jiong Zhang, Feng Zhou, Haoxue Ju, Haitao Guo, and Yuxia Du. "Research of Self-Attention in Image Segmentation." Journal of Information Technology Research 15, no. 1 (January 2022): 1–12. http://dx.doi.org/10.4018/jitr.298619.
Повний текст джерелаАнотація:
Although the traditional convolutional neural network is applied to image segmentation successfully, it has some limitations. That's the context information of the long-range on the image is not well captured. With the success of the introduction of self-attentional mechanisms in the field of natural language processing (NLP), people have tried to introduce the attention mechanism in the field of computer vision. It turns out that self-attention can really solve this long-range dependency problem. This paper is a summary on the application of self-attention to image segmentation in the past two years. And think about whether the self-attention module in this field can replace convolution operation in the future.
5
Wu, Hongqiu, Ruixue Ding, Hai Zhao, Pengjun Xie, Fei Huang, and Min Zhang. "Adversarial Self-Attention for Language Understanding." Proceedings of the AAAI Conference on Artificial Intelligence 37, no. 11 (June 26, 2023): 13727–35. http://dx.doi.org/10.1609/aaai.v37i11.26608.
Повний текст джерелаАнотація:
Deep neural models (e.g. Transformer) naturally learn spurious features, which create a ``shortcut'' between the labels and inputs, thus impairing the generalization and robustness. This paper advances self-attention mechanism to its robust variant for Transformer-based pre-trained language models (e.g. BERT). We propose Adversarial Self-Attention mechanism (ASA), which adversarially biases the attentions to effectively suppress the model reliance on features (e.g. specific keywords) and encourage its exploration of broader semantics. We conduct comprehensive evaluation across a wide range of tasks for both pre-training and fine-tuning stages. For pre-training, ASA unfolds remarkable performance gain compared to naive training for longer steps. For fine-tuning, ASA-empowered models outweigh naive models by a large margin considering both generalization and robustness.
6
Xie, Fei, Dalong Zhang, and Chengming Liu. "Global–Local Self-Attention Based Transformer for Speaker Verification." Applied Sciences 12, no. 19 (October 10, 2022): 10154. http://dx.doi.org/10.3390/app121910154.
Повний текст джерелаАнотація:
Transformer models are now widely used for speech processing tasks due to their powerful sequence modeling capabilities. Previous work determined an efficient way to model speaker embeddings using the Transformer model by combining transformers with convolutional networks. However, traditional global self-attention mechanisms lack the ability to capture local information. To alleviate these problems, we proposed a novel global–local self-attention mechanism. Instead of using local or global multi-head attention alone, this method performs local and global attention in parallel in two parallel groups to enhance local modeling and reduce computational cost. To better handle local location information, we introduced locally enhanced location encoding in the speaker verification task. The experimental results of the VoxCeleb1 test set and the VoxCeleb2 dev set demonstrated the improved effect of our proposed global–local self-attention mechanism. Compared with the Transformer-based Robust Embedding Extractor Baseline System, the proposed speaker Transformer network exhibited better performance in the speaker verification task.
7
Wang, Duofeng, Haifeng Hu, and Dihu Chen. "Transformer with sparse self‐attention mechanism for image captioning." Electronics Letters 56, no. 15 (July 2020): 764–66. http://dx.doi.org/10.1049/el.2020.0635.
Повний текст джерела
8
Li, Yujie, and Jintong Cai. "Point cloud classification network based on self-attention mechanism." Computers and Electrical Engineering 104 (December 2022): 108451. http://dx.doi.org/10.1016/j.compeleceng.2022.108451.
Повний текст джерела
9
Brotchie, James, Wei Shao, Wenchao Li, and Allison Kealy. "Leveraging Self-Attention Mechanism for Attitude Estimation in Smartphones." Sensors 22, no. 22 (November 21, 2022): 9011. http://dx.doi.org/10.3390/s22229011.
Повний текст джерелаАнотація:
Inertial attitude estimation is a crucial component of many modern systems and applications. Attitude estimation from commercial-grade inertial sensors has been the subject of an abundance of research in recent years due to the proliferation of Inertial Measurement Units (IMUs) in mobile devices, such as the smartphone. Traditional methodologies involve probabilistic, iterative-state estimation; however, these approaches do not generalise well over changing motion dynamics and environmental conditions, as they require context-specific parameter tuning. In this work, we explore novel methods for attitude estimation from low-cost inertial sensors using a self-attention-based neural network, the Attformer. This paper proposes to part ways from the traditional cycle of continuous integration algorithms, and formulate it as an optimisation problem. This approach separates itself by leveraging attention operations to learn the complex patterns and dynamics associated with inertial data, allowing for the linear complexity in the dimension of the feature vector to account for these patterns. Additionally, we look at combining traditional state-of-the-art approaches with our self-attention method. These models were evaluated on entirely unseen sequences, over a range of different activities, users and devices, and compared with a recent alternate deep learning approach, the unscented Kalman filter and the iOS CoreMotion API. The inbuilt iOS had a mean angular distance from the true attitude of 117.31∘, the GRU 21.90∘, the UKF 16.38∘, the Attformer 16.28∘ and, finally, the UKF–Attformer had mean angular distance of 10.86∘. We show that this plug-and-play solution outperforms previous approaches and generalises well across different users, devices and activities.
10
Fan, Zhongkui, and Ye-Peng Guan. "Pedestrian attribute recognition based on dual self-attention mechanism." Computer Science and Information Systems, no. 00 (2023): 16. http://dx.doi.org/10.2298/csis220815016f.
Повний текст джерелаАнотація:
Recognizing pedestrian attributes has recently obtained increasing attention due to its great potential in person re-identification, recommendation system, and other applications. Existing methods have achieved good results, but these methods do not fully utilize region information and the correlation between attributes. This paper aims at proposing a robust pedestrian attribute recognition framework. Specifically, we first propose an end-to-end framework for attribute recognition. Secondly, spatial and semantic self-attention mechanism is used for key points localization and bounding boxes generation. Finally, a hierarchical recognition strategy is proposed, the whole region is used for the global attribute recognition, and the relevant regions are used for the local attribute recognition. Experimental results on two pedestrian attribute datasets PETA and RAP show that the mean recognition accuracy reaches 84.63% and 82.70%. The heatmap analysis shows that our method can effectively improve the spatial and the semantic correlation between attributes. Compared with existing methods, it can achieve better recognition effect.
Дисертації з теми "SELF-ATTENTION MECHANISM"
1
Raykos, Bronwyn C. "Attentional and interpretive biases : independent dimensions of individual difference or expressions of a common selective processing mechanism?" University of Western Australia. School of Psychology, 2007. http://theses.library.uwa.edu.au/adt-WU2007.0018.
Повний текст джерелаАнотація:
[Truncated abstract] Attentional and interpretive biases are important dimensions of individual difference that have been implicated in the etiology and maintenance of a range of clinical problems. Yet there has been no systematic investigation into the relationship between these dimensions of individual difference. The current research program tested predictions derived from two competing theoretical accounts of the relationship between attentional and interpretive biases. The Common Mechanism Account proposes that cognitive biases represent concurrent manifestations of a single underlying selective processing mechanism. The Independent Mechanism account proposes that independent mechanisms underlie each bias. . . An apparent contradiction is that the manipulation of one bias served to also modify the other bias, despite the observation that the magnitude of the resulting change in both biases was uncorrelated. Neither the Common Mechanism nor the Independent Pathways accounts can adequately explain this pattern of results. A new account is proposed, in which attentional and interpretive biases are viewed as representing mechanisms that are related but that are not the same. Theoretical and applied implications of these findings are discussed, including the possibility that the two biases each may best predict emotional reactions to quite different stressful events and that training programs designed to attenuate allocation of attentional resources to threat may serve to reduce both attentional and interpretive selectivity in emotionally vulnerable individuals.
2
Raykos, Bronwyn C. "Attentional and interpretive biases : independent dimensions of individual difference or expressions of a common selective processing mechanism? /." Connect to this title, 2006. http://theses.library.uwa.edu.au/adt-WU2007.0018.
Повний текст джерела
3
SACHDEVA, NITIN. "CYBERBULLYING DETECTION ON SOCIAL MEDIA USING DEEP LEARNING MODELS." Thesis, DELHI TECHNOLOGICAL UNIVERSITY, 2021. http://dspace.dtu.ac.in:8080/jspui/handle/repository/18914.
Повний текст джерелаАнотація:
Application of deep learning models for cyberbullying detection in social media is an
upcoming area for both researchers and practitioners for finding, exploring and analysing
the extensibility of human-based expressions. Automated cyberbullying detection is
typically a classification problem in natural language processing where the intent is to
classify each abusive or offensive comment or post or message or image as either bullying
or non-bullying. It needs high-level semantic analysis as well. Most of the earlier attempts
on cyberbullying detection rely on manual feature extraction methods. Such methods are
not only time-consuming and cumbersome, but often fail to correctly capture the
meaning of the sentence. This fosters the need to build an intelligent analytic paradigm
for detecting cyberbullying in social media data to lower down its hazard with minimal
human intervention. Motivated by it, this research utilizes deep learning models for
cyberbullying detection in social media as they trivialize the need of explicit feature
extraction and are highly skilful, fast and more efficient in retrieval of essential features
and patterns by themselves. In our research, we have applied deep learning for
cyberbullying detection on textual and non-textual social media content. With high
volume and variety of user-generated content on complex social media platforms, the
challenges to detect cyberbullying in real-time have amplified. The influx of content
makes it challenging to timely regulate online expression. Moreover, the anonymity and
context-independence of expressions in online posts can be ambiguous or misleading.
Nowadays, cyberbullying, through varied content modalities is also very common. At the
same time, cultural diversities, unconventional use of typographical resources and easy
availability of native-language keyboards augment to the variety and volume of user-
generated content compounding the linguistic challenges in detecting online bullying
posts. In an effort to deal with this antagonistic online delinquency referred to as
cyberbullying, this research computationally analysed the content, modality and
language-use in social media using deep learning models. This research has shown that
the use of embeddings with deep learning architectures show better representation
learning capabilities and simplify the feature selection process with enhanced
classification accuracy as compared to baseline machine learning methods. The goal of
the research is to automatically detect cyberbullying on textual, multimodal and mash-up
social media content using deep learning models. In our research, we build models for
these using deep architectures including capsule network, convolution neural network,
multi-layer perceptron, self-attention mechanism, bi-directional gated recurrent unit,
long short-term memory & bi-directional long short-term memory using embeddings
such as GloVe, fastText and ELMo on social media like Askfm.in, Formspring.me, MySpace,
Twitter, YouTube, Instagram and Facebook. The results show superlative performance as
compared to SOTA as well.
4
Govindarajan, Hariprasath. "Self-Supervised Representation Learning for Content Based Image Retrieval." Thesis, Linköpings universitet, Statistik och maskininlärning, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-166223.
Повний текст джерелаАнотація:
Automotive technologies and fully autonomous driving have seen a tremendous growth in recent times and have benefitted from extensive deep learning research. State-of-the-art deep learning methods are largely supervised and require labelled data for training. However, the annotation process for image data is time-consuming and costly in terms of human efforts. It is of interest to find informative samples for labelling by Content Based Image Retrieval (CBIR). Generally, a CBIR method takes a query image as input and returns a set of images that are semantically similar to the query image. The image retrieval is achieved by transforming images to feature representations in a latent space, where it is possible to reason about image similarity in terms of image content. In this thesis, a self-supervised method is developed to learn feature representations of road scenes images. The self-supervised method learns feature representations for images by adapting intermediate convolutional features from an existing deep Convolutional Neural Network (CNN). A contrastive approach based on Noise Contrastive Estimation (NCE) is used to train the feature learning model. For complex images like road scenes where mutiple image aspects can occur simultaneously, it is important to embed all the salient image aspects in the feature representation. To achieve this, the output feature representation is obtained as an ensemble of feature embeddings which are learned by focusing on different image aspects. An attention mechanism is incorporated to encourage each ensemble member to focus on different image aspects. For comparison, a self-supervised model without attention is considered and a simple dimensionality reduction approach using SVD is treated as the baseline. The methods are evaluated on nine different evaluation datasets using CBIR performance metrics. The datasets correspond to different image aspects and concern the images at different spatial levels - global, semi-global and local. The feature representations learned by self-supervised methods are shown to perform better than the SVD approach. Taking into account that no labelled data is required for training, learning representations for road scenes images using self-supervised methods appear to be a promising direction. Usage of multiple query images to emphasize a query intention is investigated and a clear improvement in CBIR performance is observed. It is inconclusive whether the addition of an attentive mechanism impacts CBIR performance. The attention method shows some positive signs based on qualitative analysis and also performs better than other methods for one of the evaluation datasets containing a local aspect. This method for learning feature representations is promising but requires further research involving more diverse and complex image aspects.
5
Moadab, Ida. "The Role of Mindfulness and Self-Compassion in the Neural Mechanisms of Attention and Self-Monitoring." Thesis, University of Oregon, 2013. http://hdl.handle.net/1794/13402.
Повний текст джерелаАнотація:
The present study sought to investigate the effects of meditation practice on the neural mechanisms of attention and self-monitoring by comparing a group of experienced meditators to matched controls. Self-report measures of mindfulness and self-compassion were assessed to examine whether meditation-related improvements in attention and self-monitoring were linked to increases in these qualities. Thus, differences between groups (meditator versus control) on all variables and relationships among variables (attention, self-monitoring, self-compassion, and mindfulness) were explored. Results indicate that individuals with meditation experience showed enhancement in neural networks related to selective attention and attentional allocation, as evidenced by larger P1/N1 and P3b amplitudes, relative to controls. Meditators also showed improved self-monitoring of their errors, as indexed by enhanced Pe amplitudes, when compared to controls. Importantly, greater number of years of meditation experience was linked to larger Pe amplitudes, providing evidence that more practice with meditation was associated with greater error awareness. At the same time, meditators showed greater levels of mindfulness and self-compassion when compared to controls. Importantly, each of the neural indices was linked to greater levels of mindfulness and self-compassion. Specifically, self-kindness was correlated with each of these ERP components and to percentage of alpha power during meditation, and the mindfulness facet of observing fully mediated the relationship between meditation experience and P1 amplitudes. These findings suggest that the qualities that are enhanced with meditation are associated with enhancements in attentional control and awareness of errors. This study is an exciting step toward future intervention studies that combine multiple sources of information (self-report, neural measures, and behavior) to clarify the nature of the associations among these variables so that the mechanisms of mindfulness can be more fully understood.
6
Bakkali, Souhail. "Multimodal Document Understanding with Unified Vision and Language Cross-Modal Learning." Electronic Thesis or Diss., La Rochelle, 2022. http://www.theses.fr/2022LAROS046.
Повний текст джерелаАнотація:
Les modèles développés dans cette thèse sont le résultat d'un processus itératif d'analyse et de synthèse entre les théories existantes et nos études réalisées. Plus spécifiquement, nous souhaitons étudier l'apprentissage inter-modal pour la compréhension contextualisée sur les composants des documents à travers le langage et la vision. Cette thèse porte sur l'avancement de la recherche sur l'apprentissage inter-modal et apporte des contributions sur quatre fronts : (i) proposer une approche inter-modale avec des réseaux profonds pour exploiter conjointement les informations visuelles et textuelles dans un espace de représentation sémantique commun afin d'effectuer et de créer automatiquement des prédictions sur les documents multimodaux; (ii) à étudier des stratégies concurrentielles pour s'attaquer aux tâches de classification de documents intermodaux, de récupération basée sur le contenu et de classification few-shot de documents ; (iii) pour résoudre les problèmes liés aux données comme l'apprentissage lorsque les données ne sont pas annotées, en proposant un réseau qui apprend des représentations génériques à partir d'une collection de documents non étiquetés ; et (iv) à exploiter les paramètres d'apprentissage few-shot lorsque les données ne contiennent que peu d’exemplesThe frameworks developed in this thesis were the outcome of an iterative process of analysis and synthesis between existing theories and our performed studies. More specifically, we wish to study cross-modality learning for contextualized comprehension on document components across language and vision. The main idea is to leverage multimodal information from document images into a common semantic space. This thesis focuses on advancing the research on cross-modality learning and makes contributions on four fronts: (i) to proposing a cross-modal approach with deep networks to jointly leverage visual and textual information into a common semantic representation space to automatically perform and make predictions about multimodal documents (i.e., the subject matter they are about); (ii) to investigating competitive strategies to address the tasks of cross-modal document classification, content-based retrieval and few-shot document classification; (iii) to addressing data-related issues like learning when data is not annotated, by proposing a network that learns generic representations from a collection of unlabeled documents; and (iv) to exploiting few-shot learning settings when data contains only few examples
7
Yeh, Han-Yun, and 葉瀚允. "End-to-end Pinyin to Character Language Model using Self-Attention Mechanism." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/vk3zt8.
Повний текст джерелаАнотація:
碩士國立交通大學
電信工程研究所
107
Deep nerual network with conventional automatic speech recognition structure has achieved huge improvement. Similarly, end-to-end speech recongnition structure got close performance in these two years, but with huge amout of data and computing resources. This study attempt to focus on end-to-end language model, training an end-to-end language model by sequence labeling method and self-attention seq2seq model (Transformer) which are common method in some NLP task, with syllable sequence converted from 440 million words chinese corpus through a proposed G2P system. And the syllable to character model with transformer achieved lower character error rate than the baseline trigram model in our outside test set.
8
Sarzedas, João Pereira. "The impact of hallucination predisposition on vocal perception mechanisms : an ERP study of attention, identity and emotion." Master's thesis, 2020. http://hdl.handle.net/10451/45746.
Повний текст джерелаАнотація:
Tese de mestrado, Psicopatologia, Universidade de Lisboa, Faculdade de Medicina, 2020In recent years, a more dimensional view of psychotic disorders has been supported by several epidemiological studies showing the existence of a continuum of psychotic like-experiences in the general population. Reported similarities between the cognitive and neural mechanisms underpinning hallucinatory experiences in psychotic and in individuals with high hallucination predisposition provide further support for this psychosis continuum model. In the auditory domain, schizophrenic patients show altered perception of vocal information with negative content, as well as of voice identity. Whether the same voice processing mechanisms are similarly altered in individuals with high hallucination predisposition remains to be clarified. In the current study, thirty nonclinical participants with different scores on the Launay-Slade Hallucination Scale (a measure of hallucination predisposition) listened to prerecorded words varying in identity (self vs. non-self-voice) and emotional quality (negative vs. neutral vs. positive), while electroencephalographic data were recorded. Two tasks with a different attentional focus (identity focus task vs. emotional focus task) were conducted. In the identity focus task, participants attended to the identity of the voice and decided whether the voice they heard was their own voice, the voice of another person or if they were unsure. In the emotional focus task, they evaluated the emotional quality of the words. The N1, P2 and late positive potential (LPP) components were analyzed. We found a larger N1 in response to self-voice in the identity focus task and a larger LPP for the self-voice, irrespective of the attention focus. Furthermore, in the emotional focus task, we observed an association between high hallucination predisposition and a trend for a reduced LPP amplitude in response to the self-voice. The LPP result suggests the existence of an altered perception of voice identity, dependent of the attention focus, in individuals with high hallucination predisposition. Overall, the current findings partially support the psychosis continuum model.
Nas últimas décadas, uma abordagem mais dimensional das perturbações psicóticas tem ganho força com base em vários estudos epidemiológicos que demostram a existência de um continuum de experiências psicóticas na população em geral. O modelo do continuum de psicose advoga a existência de vários fenótipos de psicose num continuum que engloba indivíduos com um diagnóstico psiquiátrico (e.g., esquizofrenia) mas também indivíduos sem diagnóstico clínico que manifestam experiências do tipo psicótico. O modelo do continuum de psicose é suportado por estudos que mostram semelhanças entre os mecanismos cognitivos e neuronais que estão na base das experiências alucinatórias de pacientes psicóticos e de indivíduos sem diagnóstico clínico. As alucinações (i.e., experiências percetuais que ocorrem na ausência de estimulação externa correspondente) são um sintoma característicos das perturbações psicóticas. Contudo, também são reportadas experiências alucinatórias em pacientes com outras perturbações psiquiátricas e em indivíduos sem diagnóstico que não necessitam de apoio clínico. Os indivíduos sem diagnóstico clínico que experienciam alucinações auditivas verbais breves ou outras experiências alucinatórias são descritos na literatura como indivíduos com alta predisposição alucinatória. Numa amostra de 84711 participantes foi encontrada uma taxa de prevalência de 9.6 % de experiências alucinatórias, o que demonstra que este não é um fenómeno incomum na população em geral. A maioria das experiências alucinatórias são transitórias, mas em alguns casos persistem e aumentam o risco de transição para psicose. O estudo de experiências alucinatórias em indivíduos com predisposição alucinatória pode contribuir para uma melhor compreensão e tratamento de perturbações clínicas. Existe evidência que demonstra a existência de uma associação entre alucinações auditivas verbais e défices específicos de processamento vocal. Os modelos neurocognitivos de processamento de voz postulam que a identidade, a emoção e a fala são processados por regiões corticais funcionalmente distintas. Em indivíduos com esquizofrenia têm sido consistentemente reportadas alterações no processamento vocal da identidade, da fala e da emoção. Contudo, ainda não é claro qual o impacto destas alterações de processamento vocal no desencadear de sintomas psicóticos específicos. Examinar a existência destas alterações em indivíduos com alta predisposição alucinatória pode ser um passo útil para uma melhor compreensão da influência que estes processos têm no desencadear de experiências alucinatórias. Os escassos estudos que testaram simultaneamente o processamento de identidade, discurso e emoção parecem demonstrar diferenças de processamento vocal entre indivíduos com esquizofrenia e indivíduos com alta predisposição alucinatória. Em comparação com grupos de controlo e pacientes sem alucinações auditivas verbais, os pacientes com alucinações auditivas verbais demonstraram uma maior dificuldade no reconhecimento da sua própria voz quando esta apresentava um conteúdo negativo. Por outro lado, em indivíduos com alta predisposição alucinatória, as dificuldades no reconhecimento da própria voz não parecem estar relacionadas com o conteúdo emocional da fala. Uma das técnicas que tem vindo a demonstrar a sua utilidade para a investigação de mecanismos de perceção vocal é a técnica de potenciais evocados por eventos (event related potentials – ERP), caracterizada por elevada resolução temporal. Um dos estudos que investigou o processamento vocal da identidade, discurso e emoção com ERP encontrou efeitos de interação entre identidade (voz própria vs. voz de um desconhecido) e emoção (palavras negativas vs. neutras vs. positivas) em três componentes: N1, P2 e late positive potential (LPP). O N1 é um componente associado ao processamento sensorial da voz. O P2 é um componente associado a uma deteção precoce de saliência emocional de um estímulo. O LPP está associado a processos de atenção sustentada e de avaliação do conteúdo emocional da voz. Um estudo subsequente, utilizando um paradigma idêntico, encontrou alterações de processamento vocal, nos componentes P2 e LPP, em indivíduos com esquizofrenia. Nestes dois estudos, o conteúdo emocional foi processado implicitamente, uma vez que os participantes foram instruídos a focarem a sua atenção na identidade da voz e avaliarem se as gravações que ouviam pertenciam a eles próprios ou a outra pessoa. Alguns estudos demonstram que alterar o foco atencional de uma tarefa influencia o modo como os estímulos são processados neurofisiologicamente. Deste modo, é possível que sejam encontrados resultados diferentes numa condição onde os participantes são instruídos a focarem a sua atenção no conteúdo emocional do estímulo (i.e., avaliarem se o estímulo é negativo, neutro ou positivo). Tanto em indivíduos com esquizofrenia como em indivíduos com alta predisposição alucinatória, têm vindo a ser reportadas alterações de processos atencionais. Por conseguinte, é possível que o foco atencional module o presumível impacto da predisposição alucinatória em mecanismos de perceção de voz. No presente estudo foi explorado, através de ERPs: 1) o impacto do foco atencional na interação entre o processamento vocal da identidade e da emoção; 2) o impacto da predisposição alucinatória nestes mecanismos (foco atencional, identidade e emoção). Para esse propósito, foram realizadas duas tarefas. Em ambas as tarefas, os participantes ouviram gravações de uma voz (própria voz vs. voz de um desconhecido) a proferir palavras com conteúdo negativo, neutro ou positivo. Enquanto que na primeira tarefa (tarefa do foco na identidade) os participantes tinham de identificar a identidade da voz, na segunda tarefa (tarefa do foco na emoção) eram instruídos a avaliar o conteúdo emocional das palavras proferidas. Na tarefa de foco na identidade, procurámos replicar resultados de estudos anteriores que observaram efeitos de interação entre identidade e emoção e efeitos principais de identidade, bem como uma associação entre altos níveis de predisposição alucinatória e uma maior dificuldade no reconhecimento da própria voz. Ainda para esta tarefa, foi explorada a hipótese de que altos níveis de predisposição alucinatória predizessem um padrão neurofisiológico semelhante ao dos pacientes com esquizofrenia, uma observação que serviria de suporte para a hipótese do continuum de psicose. Na tarefa de foco na emoção era esperada uma maior amplitude de LPP em resposta a palavras negativas e positivas, em comparação com palavras neutras e, em termos exploratórios, era esperado uma influência do foco atencional no presumível impacto da predisposição alucinatória em mecanismos de perceção de voz. Foram recrutados 30 participantes “saudáveis”, sendo que o grau de predisposição alucinatória de cada individuo foi avaliado através da Launay-Slade Hallucination Scale (LSHS). No presente estudo não foram observadas interações entre identidade e emoção, nem efeitos principais de emoção. É possível que este resultado se deva a diferenças linguísticas existentes entre os estímulos usados neste estudo e os estímulos usados em estudos anteriores. Contudo, em comparação com uma voz desconhecida, a voz dos participantes elicitou uma maior amplitude de N1 na tarefa de foco na identidade e uma maior amplitude de LPP em ambas as tarefas. Estes resultados parecem sugerir que, em termos neurofisiológicos, a própria voz de uma pessoa é mais saliente do que a voz de um desconhecido. Na tarefa do foco na emoção foi também encontrada uma associação entre altos níveis de predisposição alucinatória e uma tendência para uma menor amplitude de LPP em reposta a gravações da própria voz. Este resultado sugere que o foco atencional modela o presumível impacto da predisposição alucinatória em mecanismos de perceção de voz. Ao contrário do observado em pacientes com esquizofrenia (menor amplitude de LPP em resposta a gravações da própria voz e de vozes desconhecidas com conteúdo negativo), o presumível défice de processamento vocal em indivíduos com predisposição alucinatória parece não depender do conteúdo emocional e parece ser menos marcado, uma vez que é observado apenas em resposta a gravações da própria voz e numa condição onde a identidade não está a ser explicitamente processada. Em suma, os resultados encontrados parecem servir de suporte parcial para o modelo de continuum de psicose.
Книги з теми "SELF-ATTENTION MECHANISM"
1
Mironova, Svetlana. The doctrine of the financial and legal status of municipalities in the Russian Federation and its implementation. ru: INFRA-M Academic Publishing LLC., 2021. http://dx.doi.org/10.12737/1819040.
Повний текст джерелаАнотація:
The monograph forms the doctrine of the financial and legal status of municipalities and identifies the features of its implementation in the Russian Federation at the present stage. The structure of the financial and legal status of the municipality as a whole is comprehensively presented, its elements are defined, their detailed characteristics are given; the features of the implementation of the financial and legal status of certain types of municipalities (urban and municipal districts, municipal districts, urban and rural settlements, etc.), those of them that have a special legal status (single-industry towns, BUT, science towns) or are located on the territory with a special economic status (territories of advanced socio-economic development, the free port of Vladivostok), as well as within the framework of inter-municipal cooperation and the development of urban agglomerations are indicated; a system of guarantees for the implementation by municipalities is established its financial and legal status.
Particular attention is paid to the disclosure of the social block of the financial and legal status of municipalities, represented by the mechanism of citizens' participation in the financial activities of municipalities, including the involvement of citizens in the budget process at the municipal level and the implementation of public financial control over local finances.
For researchers, teachers, postgraduates and law students, employees of public authorities and local self-government.
2
Avhadeev, V., L. Bitkova, C. Bogolyubov, I. Bondarchuk, A. Vinokurov, E. Galinovskaya, D. Gorohov, et al. Implementation of the Law on Responsible Treatment of Animals: from the quality of norms to effective law enforcement. ru: INFRA-M Academic Publishing LLC., 2021. http://dx.doi.org/10.12737/1410760.
Повний текст джерелаАнотація:
The collection contains articles on the quality of the conceptual apparatus and terminology of Federal Law No. 498-FZ of December 27, 2018 "On Responsible Treatment of Animals and on Amendments to Certain Legislative Acts of the Russian Federation", the subject of its legal regulation, the effectiveness of the mechanism for its implementation laid down in the law, state supervision and public control in the field of animal treatment. The problems of organizing the activities of animal shelters without owners, protecting animals from abuse and responsibility for such offenses, directions and ways to improve Federal Law No. 498-FZ and the practice of its application are also highlighted. Attention is paid not only to modern, but also to historical, international and foreign experience of legal regulation of the considered social relations, norms-requirements, restrictions and prohibitions in the field of keeping and using animals, moral and ethical aspects of interaction between people and animals, which emphasizes the complex and interdisciplinary nature of the presented research.
The publication is addressed to lawyers-scientists and practitioners, subjects of the law of legislative initiative, employees of state authorities and local self-government bodies directly involved in the application of the norms of Federal Law No. 498-FZ, employees of various organizations engaged in the maintenance, use and protection of animals, animal rights activists, students and postgraduates of law schools, as well as a wide range of readers interested in this issue.
3
Andrichenko, Lyudmila, A. Postnikov, L. Vasil'eva, Zh Gaunova, E. Nikitina, and Inna Plyugina. Reform of the organization of public power: the main directions of implementation. ru: INFRA-M Academic Publishing LLC., 2021. http://dx.doi.org/10.12737/1839416.
Повний текст джерелаАнотація:
The monograph examines topical issues of reforming the organization of public power in our country in connection with the adoption in 2020 of the Law on Amendments to the Constitution of the Russian Federation. The logic of changes in the organization of public power and the directions of concretization of constitutional values, taking into account the laws of the development of the constitutional system of Russia, are revealed.
The most significant characteristics of the updated model of interaction of federal public authorities in accordance with the principle of separation of powers are identified, the trends of constitutional transformations in the spheres of federal relations and local self-government, ensuring the fulfillment by public authorities of international obligations of the Russian Federation are investigated. Particular attention is paid to the development of the legal mechanism of interaction between public authorities and civil society.
The authors of the book take into account the results of legislative support for the reform of public power in 2020-2021, a forecast assessment of the implementation of the relevant constitutional and legislative novelties is given, including taking into account the existing legal risks. Solutions are proposed to a number of legal issues of legislative regulation of public power, which can increase the efficiency of its functioning.
For researchers, teachers, students and postgraduates, deputies of representative authorities, state and municipal employees, as well as anyone interested in constitutional law issues.
4
Sazhina, Muza, Anna Kashirova, Stanislav Makarov, and Egor Osiop. The social wealth of the innovation system. ru: INFRA-M Academic Publishing LLC., 2022. http://dx.doi.org/10.12737/1875920.
Повний текст джерелаАнотація:
The monograph reveals the key socio-economic problems of the innovation economy: its content as a knowledge economy and its role in evolutionary development; human capital (living intelligence) as the main resource of the innovation economy. Much attention is paid to the institutional support of innovation through a system of institutions and mutually beneficial contracts. The mixed mechanism of implementation of innovative activity as a synthesis of spontaneous market self-regulation and conscious public administration is shown. The result of the "social control" of society and the state is the coordination of the actions of economic entities and the ordering of economic processes.
The most important institution of human society is the family as a strong power in the state. And the person himself with his knowledge, culture, ethics and morality is the main value of society. The main purpose of the family is to reproduce life and provide a person with everything necessary. The state as an institution manages a person's education and health, helps to change his lifestyle, strengthening humanity, ethics, morality and culture of life.
The modern global economy remains a sphere of domination of market egoism. It is the market that performs the function of morality as a person and society as a whole. In the global economy, a person is not a representative of the people, but a representative of the system, a standard way of life. And he should live in communication based on respect for each other.
It is concluded that today the main wealth of society is not material, but social wealth: the person himself with his knowledge, culture, ethics and morality is a living intellect; a family with the reproduction of life; immaterial knowledge that covers all types of work that cannot be calculated and paid, where the motive is the joy of free cooperation, free giving and community. In this "invisible economy" people mutually teach each other humanity and create a culture of joint thinking and living together. The State and society must preserve and increase the social wealth of human society.
For students and postgraduates of economic and managerial specialties, as well as for anyone interested in this problem.
5
Crowell, Sheila E., Mona Yaptangco, and Sara L. Turner. Coercion, Invalidation, and Risk for Self-Injury and Borderline Personality Traits. Edited by Thomas J. Dishion and James Snyder. Oxford University Press, 2015. http://dx.doi.org/10.1093/oxfordhb/9780199324552.013.16.
Повний текст джерелаАнотація:
Self-inflicted injury (SII) is defined as a deliberate act in which a person seeks to cause bodily harm or death. The etiology and developmental course of SII are unclear. Converging evidence suggests coercive family processes may heighten risk for SII and related clinical problems among vulnerable youth. This chapter outlines a developmental theory of SII with particular attention to contextual risk factors. It proposes that risk for SII is highest when vulnerable youth are exposed repeatedly to coercive and invalidating family environments. Evidence in support of this theory is drawn from longitudinal studies of SII and borderline personality traits. The chapter also reviews data involving conflict discussion tasks with self-injuring and depressed adolescents and their mothers. Accumulating evidence suggests that coercive processes are a leading contextual mechanism that shapes behavioral and physiological dysregulation, ultimately heightening risk for self-injury and borderline personality disorder.
6
O'Callaghan, Claire, and Muireann Irish. Candidate Mechanisms of Spontaneous Cognition as Revealed by Dementia Syndromes. Edited by Kalina Christoff and Kieran C. R. Fox. Oxford University Press, 2018. http://dx.doi.org/10.1093/oxfordhb/9780190464745.013.6.
Повний текст джерелаАнотація:
The capacity to engage in spontaneous self-generated thought is fundamental to the human experience, yet surprisingly little is known regarding the neurocognitive mechanisms that support this complex ability. Dementia syndromes offer a unique opportunity to study how the breakdown of large-scale functional brain networks impacts spontaneous cognition. Indeed, many of the characteristic cognitive changes in dementia reflect the breakdown of foundational processes essential for discrete aspects of self-generated thought. This chapter discusses how disease-specific alterations in memory-based/construction and mentalizing processes likely disrupt specific aspects of spontaneous, self-generated thought. In doing so, it provides a comprehensive overview of the neurocognitive architecture of spontaneous cognition, paying specific attention to how this sophisticated endeavor is compromised in dementia.
7
Hasenkamp, Wendy. Catching the Wandering Mind. Edited by Kalina Christoff and Kieran C. R. Fox. Oxford University Press, 2018. http://dx.doi.org/10.1093/oxfordhb/9780190464745.013.12.
Повний текст джерелаАнотація:
This chapter considers a form of attention-based meditation as a novel means to gain insight into the mechanisms and phenomenology of spontaneous thought. Focused attention (FA) meditation involves keeping one’s attention on a chosen object, and repeatedly catching the mind when it strays from the object into spontaneous thought. This practice can thus be viewed as a kind of self-caught mind-wandering paradigm, which suggests it may have great utility for research on spontaneous thought. Current findings about the effects of meditation on mind-wandering and meta-awareness are reviewed, and implications for new research paradigms that leverage first-person reporting during FA meditation are discussed. Specifically, research recommendations are made that may enable customized analysis of individual episodes of mind-wandering and their neural correlates. It is hoped that combining detailed subjective reports from experienced meditators with rigorous objective physiological measures will advance the understanding of human consciousness.
8
Blackmore, Susan. Consciousness: A Very Short Introduction. Oxford University Press, 2017. http://dx.doi.org/10.1093/actrade/9780198794738.001.0001.
Повний текст джерелаАнотація:
Consciousness is ‘the last great mystery for science’. How can a physical brain create our experience of the world? What creates our identity? Do we really have free will? Could consciousness itself be an illusion? Exciting new developments in brain science are continuing these debates, and the field has now expanded to include biologists, neuroscientists, psychologists, and philosophers. Consciousness: A Very Short Introduction clarifies the potentially confusing arguments, and the major theories, whilst also outlining the amazing pace of neuroscience discoveries. Covering areas such as construction of self in the brain, mechanisms of attention, neural correlates of consciousness, and physiology of altered states of consciousness, it highlights the latest findings.
9
Han, Shihui. Cultural diversity. Oxford University Press, 2017. http://dx.doi.org/10.1093/acprof:oso/9780198743194.003.0001.
Повний текст джерелаАнотація:
Chapter 1 provides a brief overview of cultural differences in human behavior by giving examples of human behaviors in East Asian and Western societies. It reviews the concept of culture used by psychologists, anthropologists, and philosophers, introduces several dimensions of culture, and emphasizes shared beliefs and behavioral scripts as the key components of culture that influence human behavior. It also reviews cross-cultural psychological research that has revealed differences in multiple cognitive processes including perception, attention, memory, causal attribution, and self-reflection between individuals in East Asian and Western cultures. It gives an overview of cultural neuroscience studies that employ brain imaging techniques to reveal neural mechanisms underlying cultural differences in human behavior and mental processes.
10
Ferguson, Robert, and Karen Gillock. Memory and Attention Adaptation Training. Oxford University Press, 2021. http://dx.doi.org/10.1093/med/9780197521526.001.0001.
Повний текст джерелаАнотація:
Memory and Attention Adaptation Training (MAAT) is a cognitive-behavioral therapy (CBT) designed to help cancer survivors self-manage and mitigate the late and long-term effects of cancer and cancer therapy on memory function. Cancer-related cognitive impairment (CRCI) is a set of mild to moderate memory and attention impairments that can have an adverse influence on quality of life. CRCI symptoms tend to present during active treatment, but for some individuals cognitive changes can persist for years. While the exact prevalence of CRCI is unknown, review of the literature estimates that nearly half of all survivors may experience some form of CRCI. Causes of CRCI are multiple and are the subject of continued research. Chemotherapy, genetic vulnerability, neurovascular damage, inflammation, and hormonal/endocrine disruption have all been identified as candidate mechanisms of persistent cognitive change. Given the multiple causal mechanisms, finding a biomedical treatment for CRCI remains elusive. MAAT was developed as a CBT to help cancer survivors make adaptive behavioral and cognitive changes to improve performance in the valued activities that CRCI hinders. MAAT consists of eight visits and has been designed for administration through telehealth technology, improving access to the survivorship care that so many cancer survivors may lack after the time and expense of cancer treatment. Survivors can use this workbook to reinforce their in-session learning and continue to build adaptive coping.
Частини книг з теми "SELF-ATTENTION MECHANISM"
1
Hossain, Syed Md Minhaz, Anik Sen, and Kaushik Deb. "Detecting Spam SMS Using Self Attention Mechanism." In Intelligent Computing & Optimization, 175–84. Cham: Springer International Publishing, 2022. http://dx.doi.org/10.1007/978-3-031-19958-5_17.
Повний текст джерела
2
Daanouni, Othmane, Bouchaib Cherradi, and Amal Tmiri. "Self-Attention Mechanism for Diabetic Retinopathy Detection." In Emerging Trends in ICT for Sustainable Development, 79–88. Cham: Springer International Publishing, 2021. http://dx.doi.org/10.1007/978-3-030-53440-0_10.
Повний текст джерела
3
Zhang, Jiaxin, Yinshan Jia, and Hongfei Yu. "Application Research of YOLOv3 Incorporating Self-attention Mechanism." In Lecture Notes in Electrical Engineering, 454–61. Singapore: Springer Nature Singapore, 2022. http://dx.doi.org/10.1007/978-981-19-3927-3_45.
Повний текст джерела
4
Kanno, Satoki, and Mamoru Mimura. "Detection of Malware Using Self-Attention Mechanism and Strings." In Network and System Security, 46–60. Cham: Springer Nature Switzerland, 2023. http://dx.doi.org/10.1007/978-3-031-39828-5_3.
Повний текст джерела
5
Yuan, Genji, Jianbo Li, Yue Wang, and Xianglong Meng. "Incorporating Self Attention Mechanism into Semantic Segmentation for Lane Detection." In Wireless Algorithms, Systems, and Applications, 441–49. Cham: Springer Nature Switzerland, 2022. http://dx.doi.org/10.1007/978-3-031-19214-2_37.
Повний текст джерела
6
Cvitanović, Ivana, and Marina Bagić Babac. "Deep Learning with Self-Attention Mechanism for Fake News Detection." In Studies in Computational Intelligence, 205–29. Cham: Springer International Publishing, 2021. http://dx.doi.org/10.1007/978-3-030-90087-8_10.
Повний текст джерела
7
Zhang, Rui, Binbin Chen, and Hongbo Tang. "A Novel Sentiment Classification Architecture Based on Self-attention Mechanism." In 2020 International Conference on Data Processing Techniques and Applications for Cyber-Physical Systems, 685–92. Singapore: Springer Singapore, 2021. http://dx.doi.org/10.1007/978-981-16-1726-3_85.
Повний текст джерела
8
Li, Xiang, Ning Sun, Jixin Liu, Lei Chai, and Haian Sun. "Multi-modal Scene Recognition Based on Global Self-attention Mechanism." In Advances in Natural Computation, Fuzzy Systems and Knowledge Discovery, 109–21. Cham: Springer International Publishing, 2023. http://dx.doi.org/10.1007/978-3-031-20738-9_14.
Повний текст джерела
9
Sinh, Vu Trong, and Nguyen Le Minh. "A Study on Self-attention Mechanism for AMR-to-text Generation." In Natural Language Processing and Information Systems, 321–28. Cham: Springer International Publishing, 2019. http://dx.doi.org/10.1007/978-3-030-23281-8_27.
Повний текст джерела
10
Yu, Xu, Fayang Wan, Bin Tang, Dingjia Zhan, Qinglong Peng, Miao Yu, Zhaozhe Wang, and Shuang Cui. "Deep Bug Triage Model Based on Multi-head Self-attention Mechanism." In Computer Supported Cooperative Work and Social Computing, 107–19. Singapore: Springer Nature Singapore, 2022. http://dx.doi.org/10.1007/978-981-19-4549-6_9.
Повний текст джерелаТези доповідей конференцій з теми "SELF-ATTENTION MECHANISM"
1
Shang, Yunhao, Ning Xu, Zhenzhou Jin, and Xiao Yao. "Capsule Network Based on Self-Attention Mechanism." In 2021 13th International Conference on Wireless Communications and Signal Processing (WCSP). IEEE, 2021. http://dx.doi.org/10.1109/wcsp52459.2021.9613717.
Повний текст джерела
2
Xu, Yichong, Chenguang Zhu, Shuohang Wang, Siqi Sun, Hao Cheng, Xiaodong Liu, Jianfeng Gao, Pengcheng He, Michael Zeng, and Xuedong Huang. "Human Parity on CommonsenseQA: Augmenting Self-Attention with External Attention." In Thirty-First International Joint Conference on Artificial Intelligence {IJCAI-22}. California: International Joint Conferences on Artificial Intelligence Organization, 2022. http://dx.doi.org/10.24963/ijcai.2022/383.
Повний текст джерелаАнотація:
Most of today's AI systems focus on using self-attention mechanisms and transformer architectures on large amounts of diverse data to achieve impressive performance gains. In this paper, we propose to augment the transformer architecture with an external attention mechanism to bring external knowledge and context to bear. By integrating external information into the prediction process, we hope to reduce the need for ever-larger models and increase the democratization of AI systems. We find that the proposed external attention mechanism can significantly improve the performance of existing AI systems, allowing practitioners to easily customize foundation AI models to many diverse downstream applications. In particular, we focus on the task of Commonsense Reasoning, demonstrating that the proposed external attention mechanism can augment existing transformer models and significantly improve the model's reasoning capabilities. The proposed system, Knowledgeable External Attention for commonsense Reasoning (KEAR), reaches human parity on the open CommonsenseQA research benchmark with an accuracy of 89.4% in comparison to the human accuracy of 88.9%.
3
Pan, Yunyi, Yuan Gan, Kun Liu, and Yan Zhang. "Progressive Scene Segmentation Based on Self-Attention Mechanism." In 2020 25th International Conference on Pattern Recognition (ICPR). IEEE, 2021. http://dx.doi.org/10.1109/icpr48806.2021.9412021.
Повний текст джерела
4
Cui, Baiyun, Yingming Li, Ming Chen, and Zhongfei Zhang. "Fine-tune BERT with Sparse Self-Attention Mechanism." In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). Stroudsburg, PA, USA: Association for Computational Linguistics, 2019. http://dx.doi.org/10.18653/v1/d19-1361.
Повний текст джерела
5
Chen, Yanmin, Hao Wang, Ruijun Sun, and Enhong Chen. "Context-Aware Semantic Matching with Self Attention Mechanism." In 2022 5th International Conference on Pattern Recognition and Artificial Intelligence (PRAI). IEEE, 2022. http://dx.doi.org/10.1109/prai55851.2022.9904084.
Повний текст джерела
6
Mandhana, Vedang, and Rutuja Taware. "De novo drug design using self attention mechanism." In SAC '20: The 35th ACM/SIGAPP Symposium on Applied Computing. New York, NY, USA: ACM, 2020. http://dx.doi.org/10.1145/3341105.3374024.
Повний текст джерела
7
Xu, Song, Haoran Li, Peng Yuan, Youzheng Wu, Xiaodong He, and Bowen Zhou. "Self-Attention Guided Copy Mechanism for Abstractive Summarization." In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. Stroudsburg, PA, USA: Association for Computational Linguistics, 2020. http://dx.doi.org/10.18653/v1/2020.acl-main.125.
Повний текст джерела
8
Geng, Xinwei, Longyue Wang, Xing Wang, Bing Qin, Ting Liu, and Zhaopeng Tu. "How Does Selective Mechanism Improve Self-Attention Networks?" In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. Stroudsburg, PA, USA: Association for Computational Linguistics, 2020. http://dx.doi.org/10.18653/v1/2020.acl-main.269.
Повний текст джерела
9
Gu, Zongyun, Qing Wang, Fangfang Li, and Junling Kan. "Salient Object Detection Based on Self-attention Mechanism." In 2021 IEEE 5th Information Technology,Networking,Electronic and Automation Control Conference (ITNEC). IEEE, 2021. http://dx.doi.org/10.1109/itnec52019.2021.9587108.
Повний текст джерела
10
Li, Xinyi, Ying Qian, and Liang Dou. "BERT Named Entity Recognition with Self-attention Mechanism." In 2021 IEEE Conference on Telecommunications, Optics and Computer Science (TOCS). IEEE, 2021. http://dx.doi.org/10.1109/tocs53301.2021.9688590.
Повний текст джерелаЗвіти організацій з теми "SELF-ATTENTION MECHANISM"
1
Chornodon, Myroslava. FEAUTURES OF GENDER IN MODERN MASS MEDIA. Ivan Franko National University of Lviv, February 2021. http://dx.doi.org/10.30970/vjo.2021.49.11064.
Повний текст джерелаАнотація:
The article clarifies of gender identity stereotypes in modern media. The main gender stereotypes covered in modern mass media are analyzed and refuted. The model of gender relations in the media is reflected mainly in the stereotypical images of men and woman. The features of the use of gender concepts in modern periodicals for women and men were determined. The most frequently used derivatives of these macroconcepts were identified and analyzed in detail. It has been found that publications for women and men are full of various gender concepts that are used in different contexts. Ingeneral, theanalysisofthe concept-maximums and concept-minimum gender and their characteristics is carried out in the context of gender stereotypes that have been forme dand function in the society, system atizing the a ctual presentations. The study of the gender concept is relevant because it reveals new trends and features of modern gender images. Taking into account the special features of gender-labeled periodicals in general and the practical absence of comprehensive scientific studies of the gender concept in particular, there is a need to supplement Ukrainian science with this topic. Gender psychology, which is served by methods of various sciences, primarily sociological, pedagogical, linguistic, psychological, socio-psychological. Let us pay attention to linguistic and psycholinguistic methods in gender studies. Linguistic methods complement intelligence research tasks, associated with speech, word and text. Psycholinguistic methods used in gender psychology (semantic differential, semantic integral, semantic analysis of words and texts), aimed at studying speech messages, specific mechanisms of origin and perception, functions of speech activity in society, studying the relationship between speech messages and gender properties participants in the communication, to analyze the linguistic development in connection with the general development of the individual. Nowhere in gender practice there is the whole arsenal of psychological methods that allow you to explore psychological peculiarities of a person like observation, experiments, questionnaires, interviews, testing, modeling, etc. The methods of psychological self-diagnostics include: the gender aspect of the own socio-psychological portrait, a gender biography as a variant of the biographical method, aimed at the reconstruction of individual social experience. In the process of writing a gender autobiography, a person can understand the characteristics of his gender identity, as well as ways and means of their formation. Socio-psychological methods of studying gender include the study of socially constructed women’s and men’s roles, relationships and identities, sexual characteristics, psychological characteristics, etc. The use of gender indicators and gender approaches as a means of socio-psychological and sociological analysis broadens the subject boundaries of these disciplines and makes them the subject of study within these disciplines. And also, in the article a combination of concrete-historical, structural-typological, system-functional methods is implemented. Descriptive and comparative methods, method of typology, modeling are used. Also used is a method of content analysis for the study of gender content of modern gender-stamped journals. It was he who allowed quantitatively to identify and explore the features of the gender concept in the pages of periodicals for women and men. A combination of historical, structural-typological, system-functional methods is also implemented in the article. Descriptive and comparative methods, method of typology, modeling are used. A method of content analysis for the study of gender content of modern gender-labeled journals is also used. It allowed to identify and explore the features of the gender concept quantitatively in the periodicals for women and men. The conceptual perception and interpretation of the gender concept «woman», which is highlighted in the modern gender-labeled press in Ukraine, requires the elaboration of the polyfunctionality of gender interpretations, the comprehension of the metaphorical perception of this image and its role and purpose in society. A gendered approach to researching the gender content of contemporary periodicals for women and men. Conceptual analysis of contemporary gender-stamped publications within the gender conceptual sphere allows to identify and correlate the meta-gender and gender concepts that appear in society.