检索结果-内蒙古大学图书馆

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Hu, Xiaoyu Zhang, Xu Lin, Zexu Zhou, Deyu School of Computer Science and Engineering Southeast University Nanjing China Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications Southeast University Ministry of Education China

ISBN: (纸本)9782493814104

Code summarization is the task of automatically generating natural language descriptions from source code. Recently, pre-trained language models have gained significant popularity in code summarization due to their capacity to capture richer semantic representations of both code and natural language. Nonetheless, contemporary code summarization models grapple with two fundamental limitations. (1) Some tokens in the code are irrelevant to the natural language description and damage the alignment of the representation spaces for code and language. (2) Most approaches are based on the encoder-decoder framework, which is often plagued by the exposure bias problem, hampering the effectiveness of their decoding sampling strategies. To address the two challenges, we propose a novel pipeline framework named Reduce Redundancy then Rerank (Re3). Specifically, a redundancy reduction component is introduced to eliminate redundant information in code representation space. Moreover, a re-ranking model is incorporated to select more suitable summary candidates, alleviating the exposure bias problem. The experimental results show the effectiveness of Re3 over some state-of-the-art approaches across six different datasets from the CodeSearchNet benchmark. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Redundancy

来源：评论

学校读者我要写书评

暂无评论

Aligned Objective for Soft-Pseudo-Label Generation in Supervised Learning 41

Aligned Objective for Soft-Pseudo-Label Generation in Superv...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Xu, Ning Hu, Yihao Qiao, Congyu Geng, Xin School of Computer Science and Engineering Southeast University Nanjing China Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications Southeast University Ministry of Education China

Soft pseudo-labels, generated by the softmax predictions of the trained networks, offer a probabilistic rather than binary form, and have been shown to improve the performance of deep neural networks in supervised learning. Most previous methods adopt classification loss to train a classifier as the soft-pseudo-label generator and fail to fully exploit their potential due to the misalignment with the target of soft-pseudo-label generation, aimed at capturing the knowledge in the data rather than making definitive classifications. Nevertheless, manually designing an effective objective function for a soft-pseudo-label generator is challenging, primarily because datasets typically lack ground-truth soft labels, complicating the evaluation of the soft pseudo-label accuracy. To deal with this problem, we propose a novel framework that alternately trains the predictive model and the soft-pseudo-label generator guided by a meta-network-parameterized label enhancement objective. The parameters of the objective function are optimized based on the feedback from both the performance of the predictive model and the soft-pseudo-label generator in the learning task. Additionally, the framework offers versatility across different learning tasks by allowing direct modifications to the task loss. Experiments on the benchmark datasets validate the effectiveness of the proposed framework. Source code is available at https://***/palm-ml/SEAL. Copyright 2024 by the author(s)

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

TabLog: Test-Time Adaptation for Tabular data Using Logic Rules 41

TabLog: Test-Time Adaptation for Tabular Data Using Logic Ru...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Ren, Weijieying Li, Xiaoting Chen, Huiyuan Rakesh, Vineeth Wang, Zhuoyi Das, Mahashweta Honavar, Vasant Artificial Intelligence Research Laboratory Center for Artificial Intelligence Foundations and Scientific Applications Institute for Computational and Data Sciences College of Information Sciences and Technology The Pennsylvania State University PA United States Visa Research CA United States College of Information Sciences and Technology The Pennsylvania State University PA United States

We consider the problem of test-time adaptation of predictive models trained on tabular data. Effective solution of this problem requires adaptation of predictive models trained on the source domain to a target domain, using only unlabeled target domain data, without access to source domain data. Existing test-time adaptation methods for tabular data have difficulty coping with the heterogeneous features and their complex dependencies inherent in tabular data. To overcome these limitations, we consider test-time adaptation in the setting wherein the logical structure of the rules is assumed to remain invariant despite distribution shift between source and target domains whereas the numerical parameters associated with the rules and the weights assigned to them can vary to accommodate distribution shift. TabLog discretizes numerical features, models dependencies between heterogeneous features, introduces a novel contrastive loss for coping with distribution shift, and presents an end-to-end framework for efficient training and test-time adaptation by taking advantage of a logical neural network representation of a rule ensemble. We present results of experiments using several benchmark data sets that demonstrate TabLog is competitive with or improves upon the state-of-the-art methods for test-time adaptation of predictive models trained on tabular data. Our code is available at https://***/WeijieyingRen/TabLog. Copyright 2024 by the author(s)

关键词： data assimilation

来源：评论

学校读者我要写书评

暂无评论

Vision Transformer Based Hash Coding for Efficient Image and Audio Retrieval with Global and Local Equilibrium Distance Constraints

Vision Transformer Based Hash Coding for Efficient Image and...

引用

2024 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2024

作者： Liu, Ye Pan, Yan Yin, Jian School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou China Lizhi Inc. Artificial Intelligence and Big Data Department Guangzhou Beijing China School of Artificial Intelligence Sun Yat-sen University Zhuhai China

ISBN: (纸本)9781665410205

With the application of efficient retrieval in information systems and retrieval augmented generation with vector database for large language models, hash coding algorithms have made progress in recent years. The rise of transformer technology in the field of deep learning has brought the possibility to further improve the effect of hash coding algorithms. We introduce the vision transformer framework to both images and audios, and propose a novel approach for the tasks of multi-label image retrieval and audio event retrieval. In the proposed hash coding model, global and local equilibrium distance constraints are integrated, so that the hash codes for images can be better obtained through the global hash centers and local similar samples. In order to realize end-to-end training and hash code generation for audios, we adopt the adapter of mel spectrogram, thus the proposed approach can be simply converted and applied to audio hash coding. Comparative experiments verify that better results can be achieved on multiple image and audio datasets. © 2024 IEEE.

关键词： Image coding

来源：评论

学校读者我要写书评

暂无评论

Measuring Mutual Policy Divergence for Multi-Agent Sequential Exploration 38

Measuring Mutual Policy Divergence for Multi-Agent Sequentia...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Dou, Haowen Dang, Lujuan Luan, Zhirong Chen, Badong National Key Laboratory of Human-Machine Hybrid Augmented Intelligence China National Engineering Research Center for Visual Information and Applications China Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University China School of Electrical Engineering Xi'an University of Technology China

Despite the success of Multi-Agent Reinforcement Learning (MARL) algorithms in cooperative tasks, previous works, unfortunately, face challenges in heterogeneous scenarios since they simply disable parameter sharing for agent specialization. Sequential updating scheme was thus proposed, naturally diversifying agents by encouraging agents to learn from preceding ones. However, the exploration strategy in sequential scheme has not been investigated. Benefiting from updating one-by-one, agents have the access to the information from preceding agents. Thus, in this work, we propose to exploit the preceding information to enhance exploration and heterogeneity sequentially. We present Multi-Agent Divergence Policy Optimization (MADPO), equipped with mutual policy divergence maximization framework. We quantify the discrepancies between episodes to enhance exploration and between agents to heterogenize agents, termed intra-agent divergence and inter-agent divergence. To address the issue that traditional divergence measurements lack stability and directionality, we propose to employ the conditional Cauchy-Schwarz divergence to provide entropy-guided exploration incentives. Extensive experiments show that the proposed method outperforms state-of-the-art sequential updating approaches in two challenging multi-agent tasks with various heterogeneous scenarios. Source code is available at https://***/hwdou6677/MADPO. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multicomponent Similarity Graphs for Cross-Network Node Classification

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on artificial intelligence 2024年第3期5卷 1411-1424页

作者： Zhang, Yuhong Shi, Congmei Li, Xinzheng Zhang, Zan Hu, Xuegang School of Computer Science and Information Engineering Hefei University of Technology Hefei230601 China Institute of Artificial Intelligence Hefei Comprehensive National Science Center Hefei230088 China Key Laboratory of Knowledge Engineering with Big Data Hefei University of Technology Ministry of Education Hefei230009 China

Cross-network node classification aims to train a classifier for an unlabeled target network using a source network with rich labels. In applications, the degree of nodes mostly conforms to the long-tail distribution, i.e., most nodes in the network are tail nodes with sparse neighborhoods. The established methods focus on either the discrepancy cross network or the long tail in a single network. As for the cross-network node classification under long tail, the coexistence of sparsity of tail nodes and the discrepancy cross-network challenges existing methods for long tail or methods for the cross-network node classification. To this end, a multicomponent similarity graphs for cross-network node classification (MS-CNC) is proposed in this article. Specifically, in order to address the sparsity of the tail nodes, multiple component similarity graphs, including attribute and structure similarity graphs, are constructed for each network to enrich the neighborhoods of the tail nodes and alleviate the long-tail phenomenon. Then, multiple representations are learned from the multiple similarity graphs separately. Based on the multicomponent representations, a two-level adversarial model is designed to address the distribution difference across networks. One level is used to learn the invariant representations cross network in view of structure and attribute components separately, and the other level is used to learn the invariant representations in view of the fused structure and attribute graphs. Extensive experimental results show that the MS-CNC outperforms the state-of-the-art methods. Impact Statement-Node classification is an important task in graph mining. With the unavailability of labels, some researchers propose cross-network node classification, using one labeled network to assist the node classification of another unlabeled network. However, the long-tail of nodes leads to unsatisfactory performance and challenges the recent cross-network node classification m

关键词： Graphic methods

来源：评论

学校读者我要写书评

暂无评论

Joint Feature Representation and Classifier Learning Based Unsupervised Domain Adaption ELM

引用

Chinese Journal of Electronics 2021年第1期30卷 109-118页

作者： WANG Xuesong ZHAO Jijuan CHENG Yuhu YU Qiang School of Information and Control Engineering China University of Mining and Technology Xuzhou Key Laboratory of Artificial Intelligence and Big Data

In the problem of unsupervised domain adaption Extreme learning machine(ELM), the output layer parameters need to have both classification and domain adaptation functions, which often cannot be simultaneously fully utilized. In addition, traditional matching method based on data probability distribution cannot find the common subspace of source and target domains under large difference between domains. In order to alleviate the pressure of double functions of classifier parameters, the entire ELM learning process is mainly divided into two stages: feature representation and adaptive classifier learning, thus a joint feature representation and classifier learning based unsupervised domain adaption ELM model is proposed. In the feature representation stage, the source and target domain data are projected to their respective subspace while minimizing the difference in probability distribution between the two domains. In the adaptive classifier learning stage, the smooth manifold regularization term of target domain is used to improve the parameter adaptive *** on six different types of datasets show that the proposed model has higher cross-domain classification accuracy.

关键词： entire ELM learning process unsupervised domain adaption ELM model output layer parameters image representation double functions parameter adaptive ability common subspace classifier parameters joint feature representation unsupervised domain adaption Extreme learning machine data probability distribution learning (artificial intelligence) higher cross-domain classification accuracy target domain probability pattern classification adaptive classifier learning stage traditional matching method target domains unsupervised learning feature representation stage domain adaptation functions

来源：评论

学校读者我要写书评

暂无评论

Make Your LLM Fully Utilize the Context 38

Make Your LLM Fully Utilize the Context

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： An, Shengnan Ma, Zexiong Lin, Zeqi Zheng, Nanning Lou, Jian-Guang Chen, Weizhu National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center of Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University China Microsoft United States Peking University China

While many contemporary large language models (LLMs) can process lengthy input, they still struggle to fully utilize information within the long context, known as the lost-in-the-middle challenge. We hypothesize that it stems from insufficient explicit supervision during the long-context training, which fails to emphasize that any position in a long context can hold crucial information. Based on this intuition, our study presents INformation-INtensive (IN2) training, a purely data-driven solution to overcome lost-in-the-middle. Specifically, IN2 training leverages a synthesized long-context question-answer dataset, where the answer requires (1) fine-grained information awareness on a short segment (∼128 tokens) within a synthesized long context (4K−32K tokens), and (2) the integration and reasoning of information from two or more short segments. Through applying this information-intensive training on Mistral-7B, we present FILM-7B (FILl-in-the-Middle). To thoroughly assess the ability of FILM-7B for utilizing long contexts, we design three probing tasks that encompass various context styles (document, code, and structured-data context) and information retrieval patterns (forward, backward, and bi-directional retrieval). The probing results demonstrate that FILM-7B can robustly retrieve information from different positions in its 32K context window. Beyond these probing tasks, FILM-7B significantly improves the performance on real-world long-context tasks (e.g., 23.5→26.9 F1 score on NarrativeQA), while maintaining a comparable performance on short-context tasks (e.g., 59.3→59.2 accuracy on MMLU). © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Copula-Nested Spectral Kernel Network 41

Copula-Nested Spectral Kernel Network

引用

41st International Conference on Machine Learning, ICML 2024

作者： Tian, Jinyue Xue, Hui Xue, Yanfang Fang, Pengfei School of Computer Science and Engineering Southeast University Nanjing210096 China Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications Southeast University Ministry of Education China

Spectral Kernel Networks (SKNs) emerge as a promising approach in machine learning, melding solid theoretical foundations of spectral kernels with the representation power of hierarchical architectures. At its core, the spectral density function plays a pivotal role by revealing essential patterns in data distributions, thereby offering deep insights into the underlying framework in real-world tasks. Nevertheless, prevailing designs of spectral density often overlook the intricate interactions within data structures. This phenomenon consequently neglects expanses of the hypothesis space, thus curtailing the performance of SKNs. This paper addresses the issues through a novel approach, the Copula-Nested Spectral Kernel Network (CokeNet). Concretely, we first redefine the spectral density with the form of copulas to enhance the diversity of spectral densities. Next, the specific expression of the copula module is designed to allow the excavation of complex dependence structures. Finally, the unified kernel network is proposed by integrating the corresponding spectral kernel and the copula module. Through rigorous theoretical analysis and experimental verification, CokeNet demonstrates superior performance and significant advancements over SOTA algorithms in the field. Copyright 2024 by the author(s)

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis of Semi-refined Digital Forearm Modeling and Simplified Forearm Model in Electromagnetic Simulation

引用

Recent Patents on engineering 2024年第9期18卷 78-85页

作者： Kuang, Jiangming Qin, Yuping Zhang, Shuang School of Artificial Intelligence Neijiang Normal University Sichuan Neijiang China NJNU-OMNISKY Smart Medical Engineering Applications Joint Laboratory Neijiang Normal University Sichuan Neijiang China Neijiang City Engineering Technology Research Center of Neurological Disease Information Interference Neijiang Normal University Sichuan Neijiang China

Objective: The objective of this patent study is to analyze the performance difference between simplified and digital models based on medical images. Methods: According to the characteristics of human anatomy, the finite element simulation software COMSOL Multiphysics 5.5 was employed to construct a simplified arm model using cylinders and a digital arm model based on Chinese digital human regarding electroacupuncture therapy as an example. A comparative analysis was then performed considering three aspects: mesh number, potential distribution, and resource consumption. Results: Through analysis, the digital arm model based on Chinese digital human requires significantly more mesh cells than the simplified arm model in mesh generation. Meanwhile, because the digital arm model based on the Chinese digital human fully expresses the nonuniformity of the tissue distribution in a real human body, its signal distribution in its interior is also relatively scattered, and the coupling potential slightly differs at the electrode vertex with the smallest change. In addition, the digital arm model has much higher resource consumption and computer hardware resource requirements compared with the simplified arm model. Conclusion: As a result, the digital model based on the Chinese digital human can more fully express the tissue distribution and electrical signal characteristics of a real human body. However, due to its high computational requirements, appropriate simplification can be selected to improve the computational efficiency of the model in practical applications. © 2024 Bentham Science Publishers.

关键词： Electroacupuncture

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：