检索结果-内蒙古大学图书馆

End-to-End Tibetan Speech Recognition Study Based on Byte Pair Coding 23

学校读者我要写书评

暂无评论

23rd Chinese National Conference on Computational Linguistics, CCL 2024

作者： Cai, Yuqing Wang, Chao Duojie, Renzeng Zhu, Yulei Zhang, Jin Tashi, Nyima Tibet University School of Information Science and Technology Lhasa850000 China State Key Laboratory of Tibetan Intelligent Information Processing and Application Lhasa850000 China Qinghai Normal University College of Computer Xining810016 China

ISBN: (纸本)9780000000002

Aiming at the problems of inconsistent modeling units and unsatisfactory recognition results in Tibetan end-to-end speech recognition research, this paper proposes a BPE-Conformer-CTC/Attention end-to-end Tibetan speech recognition method. Firstly, the method adopts the byte-pair encoding algorithm for speech modeling, which solves the problem of inconsistent modeling units in Tibetan speech recognition by repeatedly merging the most frequent character pairs, segmenting the text into manageable and meaningful units, and balancing the granularity of modeling units. Secondly, the Con-former encoder is utilized to effectively fuse the global and local dependencies of audio sequences, thus enhancing the model's representational capability. Finally, the alignment and decoding process is accelerated by a joint CTC/Attention decoding strategy, which in turn improves the accuracy and efficiency of the recognition results. Experimental results on the open-source datasets XBMU-AMD031 and TIBMD@MUC show that the proposed BPE-Conformer-CTC/Attention model achieves a word error rate of 9.0% and 4.6%, respectively, and compared to the baseline Transformer-CTC/Attention model, the word error rate is reduced by 14.2% and 30.3%, respectively. This proposed approach provides an effective solution for the Tibetan end-to-end speech recognition task. © 2024 China National Conference on Computational Linguistics.

关键词： Decoding

Rethinking Centered Kernel Alignment in Knowledge Distillation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhou, Zikai Shen, Yunhang Shao, Shitong Gong, Linrui Lin, Shaohui School of Computer Science and Technology East China Normal University Shanghai China Key Laboratory of Advanced Theory and Application in Statistics and Data Science MOE China Youtu Lab Tencent Shanghai China Hunan University China

Knowledge distillation has emerged as a highly effective method for bridging the representation discrepancy between large-scale models and lightweight models. Prevalent approaches involve leveraging appropriate metrics to minimize the divergence or distance between the knowledge extracted from the teacher model and the knowledge learned by the student model. Centered Kernel Alignment (CKA) is widely used to measure representation similarity and has been applied in several knowledge distillation methods. However, these methods are complex and fail to uncover the essence of CKA, thus not answering the question of how to use CKA to achieve simple and effective distillation properly. This paper first provides a theoretical perspective to illustrate the effectiveness of CKA, which decouples CKA to the upper bound of Maximum Mean Discrepancy (MMD) and a constant term. Drawing from this, we propose a novel Relation-Centered Kernel Alignment (RCKA) framework, which practically establishes a connection between CKA and MMD. Furthermore, we dynamically customize the application of CKA based on the characteristics of each task, with less computational source yet comparable performance than the previous methods. The extensive experiments on the CIFAR-100, ImageNet-1k, and MS-COCO demonstrate that our method achieves state-of-the-art performance on almost all teacher-student pairs for image classification and object detection, validating the effectiveness of our approaches. Our code is available in https://***/Klayand/PCKA. Copyright © 2024, The Authors. All rights reserved.

关键词： Distillation

Cooperative Carrier Aircraft Support Operation Scheduling via Multi-Agent Reinforcement Learning

学校读者我要写书评

暂无评论

Cooperative Carrier Aircraft Support Operation Scheduling vi...

International Conference on Mobile Data Management

作者： Hongjie Hao Xueqin Zhang Yuan Chi Rongxin Gao Anke Xie Mingliang Xu School of Computer and Artificial Intelligence Zhengzhou University Zhengzhou China Yunnan Innovation Institute Beihang University Kunming China Yunnan Key Laboratory of Blockchain Application Technology Kunming China

Carrier Aircraft support operation is an important component of the carrier aviation support system, and its efficiency is closely related to the aircraft sortie rate. However, the carrier deck faces significant challenges such as limited space, scarce resources and constantly changing environmental conditions. These factors have a profound impact on the aircraft support operations, particularly in complex scenarios, requiring careful and rational scheduling. To solve above problems, we propose a collaborative fast scheduling framework for Multi-Agent support operations, which is based on a distributed partially observable Markov decision process (Dec-POMDP) model. Within this framework, we develop a dynamic scheduling algorithm based on centralized training with decentralized execution framework (CTDE). To evaluate the performance of our approach, we establish a simulation environment for aircraft scheduling and compare our algorithm’s performance with VDN and QMIX in various carrier aircraft scenarios. Experimental results demonstrate that our algorithm outperforms others in terms of both training efficiency and scheduling quality.

关键词：

Multi-Temporal Relationship Inference in Urban Areas

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Li, Shuangli Zhou, Jingbo Liu, Ji Xu, Tong Chen, Enhong Xiong, Hui School of Computer Science and Technology University of Science and Technology of China Baidu Research China Business Intelligence Lab Baidu Research China Big Data Lab Baidu Research China School of Computer Science and Technology University of Science and Technology of China State Key Laboratory of Cognitive Intelligence China Anhui Province Key Laboratory of Big Data Analysis and Application University of Science and Technology of China State Key Laboratory of Cognitive Intelligence China The Department of Computer Science and Engineering The Hong Kong University of Science and Technology China

Finding multiple temporal relationships among locations can benefit a bunch of urban applications, such as dynamic offline advertising and smart public transport planning. While some efforts have been made on finding static relationships among locations, little attention is focused on studying time-aware location relationships. Indeed, abundant location-based human activities are time-varying and the availability of these data enables a new paradigm for understanding the dynamic relationships in a period among connective locations. To this end, we propose to study a new problem, namely multi-Temporal relationship inference among locations (Trial for short), where the major challenge is how to integrate dynamic and geographical influence under the relationship sparsity constraint. Specifically, we propose a solution to Trial with a graph learning scheme, which includes a spatially evolving graph neural network (SEENet) with two collaborative components: spatially evolving graph convolution module (SEConv) and spatially evolving self-supervised learning strategy (SE-SSL). SEConv performs the intra-time aggregation and inter-time propagation to capture the multifaceted spatially evolving contexts from the view of location message passing. In addition, SE-SSL designs time-aware self-supervised learning tasks in a global-local manner with additional evolving constraint to enhance the location representation © 2023, CC BY.

关键词： Location

An asymptotically consistent morphoelastic shell model for compressible biological structures with finite-strain deformations

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Yu, Xiang Chen, Xiaoyi Department of Mathematics School of Computer Science and Technology Dongguan University of Technology Dongguan523808 China Guangdong Provincial Key Laboratory of Interdisciplinary Research and Application for Data Science Department of Applied Mathematics BNU-HKBU United International College Zhuhai519087 China

We derive an asymptotically consistent morphoelastic shell model to describe the finite deformations of biological tissues using the variational asymptotical method. Biological materials may exhibit remarkable compressibility when under large deformations, and we take this factor into account for accurate predictions of their morphoelastic changes. The morphoelastic shell model combines the growth model of Rodriguez et al. and a novel shell model developed by us. We start from the three-dimensional (3D) morphoelastic model and construct the optimal shell energy based on a series expansion around the middle surface. A two-step variational method is applied that retains the leading-order expansion coefficient while eliminating the higher-order ones. The main outcome is a two-dimensional (2D) shell energy depending on the stretching and bending strains of the middle surface. The derived morphoelastic shell model is asymptotically consistent with three-dimensional morphoelasticity and can recover various shell models in literature. Several examples are shown for the verification and illustration. Copyright © 2024, The Authors. All rights reserved.

关键词： Shells (structures)

PAGE: Equilibrate Personalization and Generalization in Federated Learning

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Chen, Qian Wang, Zilong Hu, Jiaqi Yan, Haonan Zhou, Jianying Lin, Xiaodong State Key Laboratory of Integrated Service Networks Xidian University Xi’an China Zhejiang Key Laboratory of Multi-dimensional Perception Technology Application and Cybersecurity Zhejiang China iTrust Singapore University of Technology and Design Singapore Singapore School of Computer Science University of Guelph Guelph Canada

Federated learning (FL) is becoming a major driving force behind machine learning as a service, where customers (clients) collaboratively benefit from shared local updates under the orchestration of the service provider (server). Representing clients’ current demands and the server’s future demand, local model personalization and global model generalization are separately investigated, as the ill-effects of data heterogeneity enforce the community to focus on one over the other. However, these two seemingly competing goals are of equal importance rather than black and white issues, and should be achieved simultaneously. In this paper, we propose the first algorithm to balance personalization and generalization on top of game theory, dubbed PAGE, which reshapes FL as a co-opetition game between clients and the server. To explore the equilibrium, PAGE further formulates the game as Markov decision processes, and leverages the reinforcement learning algorithm, which simplifies the solving complexity. Extensive experiments on four widespread datasets show that PAGE outperforms state-of-the-art FL baselines in terms of global and local prediction accuracy simultaneously, and the accuracy can be improved by up to 35.20% and 39.91%, respectively. In addition, biased variants of PAGE imply promising adaptiveness to demand shifts in practice. Copyright © 2023, The Authors. All rights reserved.

关键词： Game theory

Long-Term and Short-Term Perception in Knowledge Tracing 1

学校读者我要写书评

暂无评论

11th CCF Big Data Conference, BigData 2023

作者： Chen, Zihang Zhu, Mengxiao Wang, Fei Shen, Shuanghong Huang, Zhenya Liu, Qi Anhui Province Key Laboratory of Science Education and Communication Institute of Advanced Technology and School of Humanities and Social Sciences University of Science and Technology of China Hefei China Anhui Province Key Laboratory of Big Data Analysis and Application School of Data Science and School of Computer Science and Technology University of Science and Technology of China Hefei China

ISBN: (数字)9789819989799

ISBN: (纸本)9789819989782

Knowledge Tracing (KT) is a fundamental task in contemporary intelligent educational systems, which tracks the knowledge states of the learners based on their response sequences. KT is crucial for the effectiveness of computer-assisted intelligent educational systems, such as intelligent tutoring systems and educational resource recommendation systems. In recent years, KT models benefited from the deep learning approaches and improved dramatically compared with the traditional probabilistic approaches. However, deep learning based KT models also have significant limitations. For example, Recurrent Neural Network (RNN)-based KT models can not capture dependencies in long-term sequences effectively, and attention mechanism-based KT models rely on the positional encoding to perceive sequential information, which may disrupt the semantics of the original embeddings. Moreover, the data sparsity question remains a big challenge in existing KT models. This study is based on the observation that KT shows a stronger sequential dependence in the long term than in the short term. In this paper, we propose a novel KT model called "Long-term and Short-term perception in knowledge tracing (LSKT)", which uses multilayer perceptrons and attention mechanism to capture the long-term and short-term dependencies, and employs 2PL-IRT based embedding to alleviate the data sparsity question. Extensive experiments on multiple datasets demonstrate the effectiveness of our proposed LSKT model. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2023.

关键词： Deep Learning Educational Data Mining Intelligent Education Knowledge Tracing

LeGalFormer: A Graph Representation Learning and Transformer-based Approach for Legal Similar Case Retrieval

学校读者我要写书评

暂无评论

LeGalFormer: A Graph Representation Learning and Transformer...

International Joint Conference on Neural Networks (IJCNN)

作者： Shang Gao Yanling Li Fengpei Ge Min Lin Haiqing Yu Sukun Wang Zhongyi Miao College of Computer Science and Technology Key Laboratory of Infinite-Dimensional Hamiltonian System and Its Algorithm Application Ministry of Education Inner Mongolia Normal University Hohhot China Library Beijing University of Posts and Telecommunications Beijing China College of Computer Science and Technology Inner Mongolia Normal University Hohhot China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Legal Similar Case Retrieval (LSCR) is a critical application in legal Artificial Intelligence (AI). It involves retrieving the most relevant cases from legal case databases through query cases. Legal cases are semi-structured documents characterized by long text sequences and high specialization. Existing approaches rely on pre-trained language models for retrieval. However, these methods are constrained by the length of text input, preventing them from fully comprehending cases, which results in poor retrieval performance. Knowledge Graph (KG) are a type of graph structure data with dense knowledge and clear logic, which can represent the criminal processes and relationships among characters within legal cases. Currently, the mainstream approach for handling KG remains Graph Neural Networks (GNNs). However, these methods are limited by message passing and are prone to over-smoothing problems in the process of aggregating node features. To address these issues, we propose a Legal similar case retrieval model that combines Graph representation learning with the Transformer, called LeGalFormer. Three encoding methods are introduced to incorporate the structural information of the graph into the Transformer architecture. We evaluate the model on a real legal dataset, and the experimental results show that LeGalFormer significantly enhances the model's understanding capacity and achieves state-of-the-art performance.

关键词： Representation learning Training Law Message passing Transformers Encoding Graph neural networks

Specific Design and Implementation of the Characteristic Species Database in Wuling Mountain Area

学校读者我要写书评

暂无评论

Specific Design and Implementation of the Characteristic Spe...

Information technology, Communication Ecosystem and Management (ITCEM), International Conference on

作者： Shaowei Deng Yujun Yang Yimei Yang College of Computer and Artificial Intelligence(Software College) Huaihua University Huaihua Hunan China Key Laboratory of Wuling-Mountain Health Big Data Intelligent Processing and Application in Hunan Province Universities Huaihua Hunan China Key Laboratory of Intelligent Control Technology for Wuling-Mountain Ecological Agriculture in Hunan Province

ISBN: (数字)9798331542696

ISBN: (纸本)9798331542702

This article aims to design and implement a database system for characteristic species in the Wuling Mountain area. The system will collect, organize, store, and analyze information on characteristic species in the Wuling Mountain area, providing data support for ecological research, biodiversity conservation, agricultural production, and other fields. This article will provide a detailed introduction to database requirements analysis, data model design, database structure implementation, data security management, and system testing and optimization.

关键词： System testing Data security Ecosystems Production Database systems Data models Biodiversity Information technology Optimization