检索结果-内蒙古大学图书馆

11th International Conference on Learning Representations, ICLR 2023

作者： Li, Zenan Liu, Zehua Yao, Yuan Xu, Jingwei Chen, Taolue Ma, Xiaoxing Lü, Jian State Key Lab of Novel Software Technology Nanjing University China Department of Mathematics The University of Hong Kong Hong Kong Department of Computer Science Birkbeck University of London United Kingdom

Recent studies have explored the integration of logical knowledge into deep learning via encoding logical constraints as an additional loss ***, existing approaches tend to vacuously satisfy logical constraints through shortcuts, failing to fully exploit the *** this paper, we present a new framework for learning with logical ***, we address the shortcut satisfaction issue by introducing dual variables for logical connectives, encoding how the constraint is *** further propose a variational framework where the encoded logical constraint is expressed as a distributional loss that is compatible with the model's original training *** theoretical analysis shows that the proposed approach bears salient properties, and the experimental evaluations demonstrate its superior performance in both model generalizability and constraint satisfaction. © 2023 11th International Conference on Learning Representations, ICLR 2023. All rights reserved.

关键词： Signal encoding

来源：评论

学校读者我要写书评

暂无评论

Revisiting Stochastic Multi-Level Compositional Optimization

引用

IEEE Transactions on Pattern Analysis and Machine Intelligence 2025年第7期47卷 5613-5624页

作者： Wei Jiang Sifan Yang Yibo Wang Tianbao Yang Lijun Zhang National Key Laboratory for Novel Software Technology Nanjing University Nanjing China National Key Laboratory for Novel Software Technology School of Artiicial Intelligence Nanjing University Nanjing China Department of Computer Science and Engineering Texas A&M University College Station TX USA

This paper explores stochastic multi-level compositional optimization, where the objective function is a composition of multiple smooth functions. Traditional methods for solving this problem suffer from either sub-optimal sample complexities or require huge batch sizes. To address these limitations, we introduce the Stochastic Multi-level Variance Reduction (SMVR) method. In the expectation case, our SMVR method attains the optimal sample complexity of $\mathcal {O}(1/\epsilon ^{3})$ to find an $\epsilon$ -stationary point for non-convex objectives. When the function satisfies convexity or the Polyak-Łojasiewicz (PL) condition, we propose a stage-wise SMVR variant. This variant improves the sample complexity to $\mathcal {O}(1/\epsilon ^{2})$ for convex functions and $\mathcal {O}(1/(\mu \epsilon ))$ for functions meeting the $\mu$ -PL condition or $\mu$ -strong convexity. These complexities match the lower bounds not only in terms of $\epsilon$ but also in terms of $\mu$ (for PL or strongly convex functions), without relying on large batch sizes in each iteration. Furthermore, in the finite-sum case, we develop the SMVR-FS algorithm, which can achieve a complexity of $\mathcal {O}(\sqrt{n}/\epsilon ^{2})$ for non-convex objectives, $\mathcal {O}(\sqrt{n}/\epsilon \log (1/\epsilon ))$ for convex functions and $\mathcal {O}(\sqrt{n}/\mu \log (1/\epsilon ))$ for objectives satisfying the $\mu$ -PL condition, where $n$ denotes the number of functions in each level. To make use of adaptive learning rates, we propose the Adaptive SMVR method, which maintains the same complexities while demonstrating faster convergence in practice.

关键词： Complexity theory Optimization Convex functions Convergence Training Stochastic processes Lower bound Jacobian matrices Adaptive learning Storms

来源：评论

学校读者我要写书评

暂无评论

On Efficient Packet Batching and Resource Allocation for GPU based NFV Acceleration 31

On Efficient Packet Batching and Resource Allocation for GPU...

引用

31st IEEE/ACM International Symposium on Quality of Service, IWQoS 2023

作者： Zeng, Deze Zhu, Andong Gu, Lin Chen, Quan Guo, Minyi School of Computer Science China University of Geosciences Hubei Wuhan China Nanjing University State Key Lab. for Novel Software Technology Nanjing China School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China Shanghai Jiao Tong University Department of Computer Science and Engineering Shanghai China

ISBN: (纸本)9798350399738

Network Function Virtualization (NFV) has already become an essential technology for improving the scalability and flexibility of modern computer networks. The performance gap has become the main issue that impedes the development of NFV. GPUs, with massive parallel processors, are advocated to accelerate the Virtualized Network Functions (VNFs). However, the special architecture and workflow of GPUs introduce new challenges, especially on the batched processing, and resource allocation. In this paper, we propose GPU-based NFV Acceleration framework (GNFA) with an efficient packet batching and resource allocation solution. Considering the increased latency caused by the accumulation of the GPU kernel invoking overhead, we first invent a latency reduction mechanism called SM Performance Compensation (SPC). A Partition and Adjustment based Batching and Resource Allocation (PABARA) algorithm that jointly considers batch size tuning and GPU thread allocation is also proposed. We have practically implemented GNFA and extensively evaluated its performance on some well-known VNFs. The experiment results show that GNFA can effectively promote the GPU resource utilization and improve the NFV performance in terms of per-packet latency. © 2023 IEEE.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

Hashing based Contrastive Learning for Virtual Screening

arXiv

引用

arXiv 2024年

作者： Han, Jin Hong, Yun Li, Wu-Jun National Key Laboratory for Novel Software Technology Department of Computer Science and Technology Nanjing University Nanjing210023 China

Virtual screening (VS) is a critical step in computer-aided drug discovery, aiming to identify molecules that bind to a specific target receptor like protein. Traditional VS methods, such as docking, are often too time-consuming for screening large-scale molecular databases. Recent advances in deep learning have demonstrated that learning vector representations for both proteins and molecules using contrastive learning can outperform traditional docking methods. However, given that target databases often contain billions of molecules, real-valued vector representations adopted by existing methods can still incur significant memory and time costs in VS. To address this problem, in this paper we propose a hashing-based contrastive learning method, called DrugHash, for VS. DrugHash treats VS as a retrieval task that uses efficient binary hash codes for retrieval. In particular, DrugHash designs a simple yet effective hashing strategy to enable end-to-end learning of binary hash codes for both protein and molecule modalities, which can dramatically reduce the memory and time costs with higher accuracy compared with existing methods. Experimental results show that DrugHash can outperform existing methods to achieve state-of-the-art accuracy, with a memory saving of 32× and a speed improvement of 3.5×. Copyright © 2024, The Authors. All rights reserved.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

A Privacy Preserving Method for IoT Forensics

A Privacy Preserving Method for IoT Forensics

引用

2024 IEEE Global Communications Conference, GLOBECOM 2024

作者： Zhang, Wenzheng Chen, Boxi Fu, Xiao Gu, Qing Shi, Jin Du, Xiaojiang Zhou, Xiaoyang Nanjing University State Key Laboratory for Novel Software Technology Nanjing China Nanjing University School of Information Management Nanjing China Stevens Institute of Technology Department of Electrical and Computer Engineering HobokenNJ United States Innovation Research Institute Nanjing China

ISBN: (纸本)9798350351255

The diversity of the Internet of Things (IoT) poses challenges to privacy protection, especially in the field of digital forensics. How to ensure that only the private information of the suspect is provided, and not the irrelevant information of other users is disclosed is crucial, especially when obtaining evidence in the complex IoT environment. To the best of our knowledge, there are few studies on protecting the privacy of irrelevant users in the IoT forensics. However, it is very important to ensure that the evidence does not violate the privacy of other users when collecting evidence, because it directly determines whether the evidence is legal and whether it can be admissible in court. In this paper, a new method based on data provenance graph is designed to solve the privacy protection problem of IoT forensics. The key idea of this method is to protect privacy by dividing multi-user information and protecting it from an encryption perspective. The method consists of three main phrases: distinguishing different users' data provenance graphs using traversal search, node abstraction, and hiding techniques, utilizing pseudo-random dual-key negotiation methods tailored for the scenario to enhance privacy protection for unrelated users, and employing identity authentication technology to facilitate better investigation and extraction of data provenance graph information of criminal accomplices in specific scenarios. Example proves that this method has practical significance and promising application prospects in protecting the privacy of unrelated users in IoT forensics while ensuring evidence accessibility in special criminal scenarios. © 2024 IEEE.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

Hashing for Protein Structure Similarity Search

arXiv

引用

arXiv 2024年

作者： Han, Jin Li, Wu-Jun National Key Laboratory for Novel Software Technology Department of Computer Science and Technology Nanjing University Nanjing210023 China

Protein structure similarity search (PSSS), which tries to search proteins with similar structures, plays a crucial role across diverse domains from drug design to protein function prediction and molecular evolution. Traditional alignment-based PSSS methods, which directly calculate alignment on the protein structures, are highly time-consuming with high memory cost. Recently, alignment-free methods, which represent protein structures as fixed-length real-valued vectors, are proposed for PSSS. Although these methods have lower time and memory cost than alignment-based methods, their time and memory cost is still too high for large-scale PSSS, and their accuracy is unsatisfactory. In this paper, we propose a novel method, called protein structure hashing (POSH), for PSSS. POSH learns a binary vector representation for each protein structure, which can dramatically reduce the time and memory cost for PSSS compared with real-valued vector representation based methods. Furthermore, in POSH we also propose expressive hand-crafted features and a structure encoder to well model both node and edge interactions in proteins. Experimental results on real datasets show that POSH can outperform other methods to achieve state-of-the-art accuracy. Furthermore, POSH achieves a memory saving of more than six times and speed improvement of more than four times, compared with other methods. © 2024, CC BY.

关键词： Molecular docking

来源：评论

学校读者我要写书评

暂无评论

Testing Constraint Checking Implementations via Principled Metamorphic Transformations

Testing Constraint Checking Implementations via Principled M...

引用

IEEE International Conference on software Analysis, Evolution and Reengineering (SANER)

作者： Mingchen Gao Huiyan Wang Chang Xu State Key Laboratory for Novel Software Technology Nanjing University Nanjing China Department of Computer Science and Technology Nanjing University Nanjing China

ISBN: (数字)9798350330663

ISBN: (纸本)9798350330670

Constraint checking techniques are being widely used for ensuring the consistency of software artifacts during their development and evolution (e.g., detecting inconsistency in an application's running contexts or identifying rule violation in the code being developed). Typically, consistency constraints are formulated and checked upon the changes of software artifacts under checking. When any constraint is violated, an inconsistency is said to occur and then follow-up actions can be taken to remedy the problem. Currently, various constraint checking techniques have been proposed and implemented with sophisticated mechanisms for higher efficiency and scalability. However, these implementations could be far from being well tested due to their oracle problems, i.e., hardly able to tell what the checking result should be, given any software artifacts and their consistency constraints to check. In this paper, we leverage metamorphic testing and propose a family of metamorphic relations catered for testing constraint checking implementations. We dedicatedly design these relations by following the sensitivity principle via a fine-granularity control and the diversity principle via input-oriented transformations. Our experiments reported promising results (disclosing 80 % mutation bugs and five real bugs) without the need of any manual labeling.

关键词： Sensitivity Codes Scalability computer bugs Manuals Fuzzing software systems

来源：评论

学校读者我要写书评

暂无评论

FedLPS: Heterogeneous Federated Learning for Multiple Tasks with Local Parameter Sharing

arXiv

引用

arXiv 2024年

作者： Jia, Yongzhe Zhang, Xuyun Beheshti, Amin Dou, Wanchun State Key Laboratory for Novel Software Technology Department of Computer Science and Technology Nanjing University China School of Computing Macquarie University Australia

Federated Learning (FL) has emerged as a promising solution in Edge Computing (EC) environments to process the proliferation of data generated by edge devices. By collaboratively optimizing the global machine learning models on distributed edge devices, FL circumvents the need for transmitting raw data and enhances user privacy. Despite practical successes, FL still confronts significant challenges including constrained edge device resources, multiple tasks deployment, and data heterogeneity. However, existing studies focus on mitigating the FL training costs of each single task whereas neglecting the resource consumption across multiple tasks in heterogeneous FL scenarios. In this paper, we propose heterogeneous FEDerated learning with Local Parameter Sharing (FedLPS) to fill this gap. FedLPS leverages principles from transfer learning to facilitate the deployment of multiple tasks on a single device by dividing the local model into a shareable encoder and task-specific predictors. To further reduce resource consumption, a channel-wise model pruning algorithm that shrinks the footprint of local models while accounting for both data and system heterogeneity is employed in FedLPS. Additionally, a novel heterogeneous model aggregation algorithm is proposed to aggregate the heterogeneous predictors in FedLPS. We implemented the proposed FedLPS on a real FL platform and compared it with state-of-the-art (SOTA) FL frameworks. The experimental results on five popular datasets and two modern DNN models illustrate that the proposed FedLPS significantly outperforms the SOTA FL frameworks by up to 4.88% and reduces the computational resource consumption by 21.3%. Our code is available at: https://***/jyzgh/FedLPS. © 2024, CC BY.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Jianqing Song Jianguo Huang Wenyu Jiang Baoming Zhang Shuangjie Li Chongjun Wang State Key Laboratory of Novel Software Technology Nanjing University Department of Statistics and Data Science Southern University of Science and Technology and College of Computing and Data Science Nanyang Technological University Department of Statistics and Data Science Southern University of Science and Technology and State Key Laboratory of Novel Software Technology Nanjing University

ISBN: (纸本)9798331314385

Graph Neural Networks have achieved remarkable accuracy in semi-supervised node classification tasks. However, these results lack reliable uncertainty estimates. Conformal prediction methods provide a theoretical guarantee for node classification tasks, ensuring that the conformal prediction set contains the ground-truth label with a desired probability (e.g., 95%). In this paper, we empirically show that for each node, aggregating the non-conformity scores of nodes with the same label can improve the efficiency of conformal prediction sets while maintaining valid marginal coverage. This observation motivates us to propose a novel algorithm named Similarity-Navigated Adaptive Prediction Sets (SNAPS), which aggregates the non-conformity scores based on feature similarity and structural neighborhood. The key idea behind SNAPS is that nodes with high feature similarity or direct connections tend to have the same label. By incorporating adaptive similar nodes information, SNAPS can generate compact prediction sets and increase the singleton hit ratio (correct prediction sets of size one). Moreover, we theoretically provide a finite-sample coverage guarantee of SNAPS. Extensive experiments demonstrate the superiority of SNAPS, improving the efficiency of prediction sets and singleton hit ratio while maintaining valid coverage.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Distributed GAN: Toward a Faster Reinforcement-Learning-Based Architecture Search

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2022年第3期3卷 391-401页

作者： Shi, Jiachen Fan, Yi Zhou, Guoqiang Shen, Jun Nanjing University of Posts and Telecommunications College of Computer Science Nanjing210046 China Nanjing University State Key Laboratory for Novel Software Technology Nanjing210046 China University of Wollongong School of Computing and Information Technology WollongongNSW2522 Australia

In the existing reinforcement learning (RL)-based neural architecture search (NAS) methods for a generative adversarial network (GAN), both the generator and the discriminator architecture are usually treated as the search objects. In this article, we take a different perspective to propose an approach by treating the generator as the search objective and the discriminator as the judge to evaluate the performance of the generator architecture. Consequently, we can convert this NAS problem to a GAN-style problem, similar to using a controller to generate sequential data via reinforcement learning in a sequence GAN, except that the controller in our methods generates serialized data information of architecture. Furthermore, we adopt an RL-based distributed search method to update the controller parameters θ. Generally, the reward value is calculated after the whole architecture searched, but as another novelty in this article, we employ the reward shaping method to judge the intermediate reward and assign it to every cell in the architecture to encourage the diversity and the integrity of all cells. The main contribution of this article is to provide a novel performance estimation mechanism, which could speed up the efficiency of architecture search, and improve the searching results with specific supplementary strategies. Crucially, this estimation mechanism can be applied to most RL-based NAS methods for the GAN. The experiments demonstrate that our methods achieve satisfactory results against our design objectives. © 2020 IEEE.

关键词： Reinforcement learning

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：