检索结果-内蒙古大学图书馆

作者： Sim, Jae-Young Lee, Sang-Uk IEEE Computing and Intelligence Laboratory Samsung Advanced Institute of Technology Yongin Korea Republic of School of Electrical Engineering and Computer Science Seoul National University Seoul 151-742 Korea Republic of

In this paper, we propose adaptive and flexible quantization and compression algorithms for 3-D point data using vector quantization (VQ) and rate-distortion (R-D) optimization. The point data are composed of the position and the radius of sphere based on QSplat representation. The positions of child spheres are first transformed to the local coordinate system, which is determined by the parentchildren relationship. The local coordinate transform makes the positions more compactly distributed in 3-D space, facilitating an effective application of VQ. We also develop a constrained encoding method for the radius data, which can provide a hole-free surface rendering at the decoder side. Furthermore, R-D optimized compression algorithm is proposed in order to allocate an optimal bitrate to each sphere. Experimental results show that the proposed algorithm can effectively compress the original 3-D point geometry at various bitrates. © 2006 IEEE.

关键词： Data visualization

来源：评论

学校读者我要写书评

暂无评论

Joint Feature Fusion Approach to Human Pose Estimation Using mmWave Radar 5th

Joint Feature Fusion Approach to Human Pose Estimation Usin...

引用

5th China Conference on Intelligent Networked Things, CINT 2022

作者： Zhang, Jianxiong Cao, Zhongping Ding, Wen Chen, Rihui Guo, Xuemei Wang, Guoli School of Artificial Intelligence Sun-Yat-Sen University Zhuhai519000 China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Beijing China School of Computer Science and Engineering Sun-Yat-Sen University Guangzhou510006 China

ISBN: (纸本)9789811989148

This paper mainly studies the use of millimeter wave (mmWave) radar for 3D human pose estimation. Although pioneering works have achieved remarkable success, the lack of stability of continuous data acquisition by mmWave radar is still an intractable problem in pose capturing. To mitigate this problem, a joint feature fusion method is proposed, which is based on the assumption that human pose in a motion sequence is domained by bone connection and specific motion cooperation mechanism. Specifically, we introduce a forget gate to suppress error features extracted due to the loss of information, and a select gate to select the most reasonable joint characteristics. Outputs of the two gates are fused to form an anti distortion mechanism of human pose estimation. Experiments show the effectiveness of our method. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Data acquisition

来源：评论

学校读者我要写书评

暂无评论

Improving Visual Quality of Image Synthesis by A Token-based Generator with Transformers 35

Improving Visual Quality of Image Synthesis by A Token-based...

引用

35th Conference on Neural Information Processing Systems, NeurIPS 2021

作者： Zeng, Yanhong Yang, Huan Wang, Jianbo Fu, Jianlong School of Computer Science and Engineering Sun Yat-sen University China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China Microsoft Research Asia China University of Tokyo Japan

ISBN: (纸本)9781713845393

We present a new perspective of achieving image synthesis by viewing this task as a visual token generation problem. Different from existing paradigms that directly synthesize a full image from a single input (e.g., a latent code), the new formulation enables a flexible local manipulation for different image regions, which makes it possible to learn content-aware and fine-grained style control for image synthesis. Specifically, it takes as input a sequence of latent tokens to predict the visual tokens for synthesizing an image. Under this perspective, we propose a token-based generator (i.e.,TokenGAN). Particularly, the TokenGAN inputs two semantically different visual tokens, i.e., the learned constant content tokens and the style tokens from the latent space. Given a sequence of style tokens, the TokenGAN is able to control the image synthesis by assigning the styles to the content tokens by attention mechanism with a Transformer. We conduct extensive experiments and show that the proposed TokenGAN has achieved state-of-the-art results on several widelyused image synthesis benchmarks, including FFHQ and LSUN CHURCH with different resolutions. In particular, the generator is able to synthesize high-fidelity images with 1024 × 1024 size, dispensing with convolutions entirely. © 2021 Neural information processing systems foundation. All rights reserved.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Network Embedding with Topology-Aware Textual Representations 18th

Network Embedding with Topology-Aware Textual Representation...

引用

18th Pacific Rim International Conference on Artificial intelligence, PRICAI 2021

作者： Chen, Jiaxing Xu, Zenan Su, Qinliang School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Guangzhou China

ISBN: (纸本)9783030891879

Textual network embedding aims to learn meaningful low-dimensional representations for vertices with the consideration of the associated texts. When learning the representations for texts in network embedding, existing methods mostly only exploit information from neighboring texts (i.e., contexts), while rarely taking advantages of the valuable network topological (structural) information. To bridge the gap, in this paper, a model based on adaptive-filter convolutional neural networks (CNN) is developed, in which the filters are adapted to local network topologies, rather than clamped to fixed values as in traditional CNNs. The dependency enables the learned text representations to be aware of local network topologies. It is shown that the proposed topology-aware representations can be viewed as a complement to existing context-aware ones. When the two are used together, experimental results on three real-world benchmarks demonstrate that significant performance improvements on the tasks of link prediction and vertex classification. © 2021, Springer Nature Switzerland AG.

关键词： Network topology

来源：评论

学校读者我要写书评

暂无评论

Deadline constrained cloud computing resources scheduling for cost optimization based on dynamic objective genetic algorithm

Deadline constrained cloud computing resources scheduling fo...

引用

IEEE Congress on Evolutionary Computation, CEC 2015

作者： Chen, Zong-Gan Du, Ke-Jing Zhan, Zhi-Hui Zhang, Jun Department of Computer Science Sun Yat-Sen University Guangzhou510275 China School of Advanced Computing Sun Yat-Sen University Guangzhou510275 China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China Engineering Research Center of Supercomputing Engineering Software Ministry of Education China Key Laboratory of Software Technology Department of Guangdong Province China

ISBN: (纸本)9781479974924

Cloud computing resources scheduling is significant for executing the workflows in cloud platform because it relates to both the execution time and execution cost. In order to take both the time and cost into consideration, Rodriguez and Buyya have proposed a cost-minimization and deadline-constrained workflow scheduling model on cloud computing. Their model has great applicability but the solution of their particle swarm optimization (PSO) approach is not good enough and cannot meet a tight deadline condition. In this paper, we propose a genetic algorithm (GA) approach to solve this model. In order to tackle with the tight deadline condition, a dynamic objective strategy is further proposed to let GA focus on optimize the execution time objective to meet the deadline constraint when the feasible solution hasn't been obtained. After obtaining a feasible solution, the GA focuses on optimizing the execution cost within the deadline constraint. Therefore, the proposed dynamic objective GA (DOGA) has adaptive ability to the search environment to different objectives. We have conduct extensive experiments based on workflows with different scales and different cloud resources. Experimental results show that DOGA can find better solution with smaller cost than PSO does on different scheduling scales and different deadline conditions. DOGA approach is more applicable to be used in commercial activities. © 2015 IEEE.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Using Knowledge Induction Strategies: Llms Can Do Better in Knowledge-Driven Dialogue Tasks

SSRN

引用

SSRN 2023年

作者： Peng, Sisi Qu, Dan Zhang, Hao Zhang, Wenlin Li, Shunhang Information Engineering University Zhengzhou China Laboratory for Advanced Computing and Intelligence Engineering Zhengzhou China

Acquiring and utilizing knowledge is essential for dialogue systems to generate high-quality responses. In recent years, Large Language Models (LLMs) have garnered significant attention in research due to their ability to learn general knowledge from vast amounts of training data and complex textual semantics using deep neural network structures. However, LLMs face certain limitations in knowledge-driven dialogue tasks, including a lack of factual knowledge and poor utilization of commonsense knowledge. Many efforts alleviate these issues by providing an external knowledge base, which is time-consuming and labor-intensive. To address these shortcomings, this paper introduces two Knowledge Induction (KI) strategies: Explicit Knowledge Induction (EKI) and Implicit Knowledge Induction (IKI), to mine LLMs’ internel knowledge and stimulate their own reasoning ability. EKI involves an iterative automatic prompt generation process that guides LLMs in generating explicit knowledge, while IKI utilizes Chain-Of-Thought (CoT) prompts to construct dialogue thinking chains. These strategies empower LLMs to accurately and comprehensively leverage internal knowledge, thereby enabling them to tackle knowledge-driven dialogue tasks without relying on external knowledge bases. Furthermore, the detailed conditional constraints embedded in the prompts facilitate the control of knowledge generation formats, streamlining the knowledge post-processing process. To validate the efficacy of the KI strategies, a series of experiments were performed across multiple models and public datasets. The experimental results demonstrate that the KI strategies effectively enhance knowledge utilization and improve the quality of dialogue generation. Specifically, ChatGLM achieved significant improvements in BLEU (+3.86/+0.83), ROUGE (+7.55/+7.94), and MAUVE (+0.1/+0.08) metrics. © 2023, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Easy and Effective! Data Augmentation for Knowledge-Aware Dialogue Generation Via Multi-Perspective Sentences Interaction

SSRN

引用

SSRN 2024年

作者： Peng, Sisi Qu, Dan Zhang, Wenlin Zhang, Hao Li, Shunhang Xu, Minchen Information Engineering University Zhengzhou China Laboratory for Advanced Computing and Intelligence Engineering Zhengzhou China

In recent years, knowledge-based dialogue generation has become a popular research topic due to its ability to generate informative and coherent responses by incorporating external knowledge into models. However, acquiring high-quality knowledge that corresponds to the dialogue content can be difficult and requires significant time and resources. To address the issue of limited dialogue data, most research efforts focus on data augmentation to obtain more training data. Unfortunately, these approaches do not consider knowledge augmentation, resulting in limited diversity of input data and only achieving improvements on a single metric. Real-life conversations are diverse, with repetitions, reversals, and interruptions, which present a higher demand for data diversity. In this paper, we propose a simple and effective data augmentation method called Multi-perspective Sentence Interaction to enhance the connections between sentences from different perspectives. By analyzing the target responses from multiple dialogue perspectives, we gain a better understanding of the relationships between dialogue sentences, thereby achieving an expanded knowledge-based dialogue data. We conduct experiments on multiple knowledge-based dialogue datasets and different models, and the results demonstrate that our method significantly improves the quality of model generation. Specifically, we achieved a 3.5% increase in reply accuracy and a 0.1506 improvement in diversity (DIST-2). Furthermore, the accuracy of knowledge selection improved by 19.04%, and the model perplexity decreased by 31.48%. © 2024, The Authors. All rights reserved.

关键词： Knowledge based systems

来源：评论

学校读者我要写书评

暂无评论

Exploiting Multiple Features for Hash Codes Learning with Semantic-Alignment-Promoting Variational Auto-encoder 12th

Exploiting Multiple Features for Hash Codes Learning with Se...

引用

12th National CCF Conference on Natural Language Processing and Chinese computing, NLPCC 2023

作者： Chen, Jiayang Su, Qinliang School of Computer Science and Engineering Sun Yat-Sen University Guangzhou China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Beijing China

ISBN: (纸本)9783031446924

Semantic hashing is an effective technique to empower information retrieval. Currently, considerable efforts have been dedicated to generating high-quality hash codes by modeling document features using generative models and other approaches. However, most of these methods rely solely on a single type of feature, such as TFIDF features, BERT embeddings, etc. As different types of features have distinct but complementary information of documents, e.g. TFIDF mainly contains the keywords information and BERT focuses on the semantics, hash codes generated solely from either may not capture the full essence of the documents. To overcome this challenge, we propose a semantic-alignment-promoting variational auto-encoder to generate hash codes from multiple document features. Specifically, a VAE-based generative model is first developed to model the multiple features. Then, we propose a semantic-alignment-promoting inference network to estimate the parameters of the variational posterior from multiple features. Additionally, the quality of hash codes is further improved by promoting the semantic alignment between the hash codes of connected documents in a constructed connection graph. The results of extensive experiments on three public datasets demonstrate that our proposed model significantly outperforms current state-of-the-art models. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Meta-Adaptable-Adapter: Efficient Adaptation of Self-Supervised Models for Low-Resource Speech Recognition

SSRN

引用

SSRN 2023年

作者： Chen, Yaqi Zhang, Hao Yang, Xukui Zhang, Wenlin Qu, Dan Information Engineering University Zhengzhou China Laboratory for Advanced Computing and Intelligence Engineering Zhengzhou China

Self-supervised models have demonstrated remarkable performance in speech processing by learning latent representations from large amounts of unlabeled data. Adapting these models to low-resource languages yields promising results, but the computational cost of fine-tuning all model parameters is prohibitively high. Adapters offer a solution by introducing lightweight bottleneck structures into pre-trained models for downstream tasks, enabling efficient parameter adaptation. However, randomly initialized adapters often underperform in extremely low-resource scenarios. To address this issue, we explore the Meta-Adapter for self-supervised models and analyzed some limitations of Meta-Adapter including poor learning in language-specific knowledge and meta-overfitting problems. To relieve these problems, we propose the Meta-Adaptable-Adapter (MAA), a new meta leaning algorithm that adapts to low-resource languages quickly and effectively. MAA learns task-specific adapters for feature extraction, and task-independent adapters for feature combination. The experiments on the Common Voice and FLEURS datasets show superior performance on sixteen low-resource languages across four different language families compared to other adapters, showing better generalization and extensibility. © 2023, The Authors. All rights reserved.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

A pso approach for software project planning 14

A pso approach for software project planning

引用

16th Genetic and Evolutionary Computation Conference Companion, GECCO 2014 Companion

作者： Jia, Ya-Hui Chen, Wei-Neng Hu, Xiao-Min Department of Computer Science Sun Yat-sen University China School of Advanced Computing Sun Yat-sen University China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China Engineering Research Center of Supercomputing Engineering Software Ministry of Education China School of Public Health Sun Yat-sen University China

ISBN: (纸本)9781450328814

Search-based software project management is a hot research point in software engineering. Based on the event-based scheduler (EBS) we have proposed in previous work [1], this paper intends to further propose a two-phase particle swarm optimization approach which uses a set-based representation for task scheduling and an integer representation for workload assignment scheduling to improve planning performance. Experimental results on 83 instances demonstrate the effectiveness of the proposed approach.

关键词： Scheduling

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：