检索结果-内蒙古大学图书馆

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Dan, Jun Liu, Weiming Xie, Chunfeng Yu, Hua Dong, Shunjie Tan, Yanchao Zhejiang University China Queen Mary University of London United Kingdom Dalian University of Technology China Shanghai Jiao Tong University China Fuzhou University China Engineering Research Center of Big Data Intelligence Ministry of Education China Fujian Key Laboratory of Network Computing and Intelligent Information Processing China

Semi-supervised graph domain adaptation, as a branch of graph transfer learning, aims to annotate unlabeled target graph nodes by utilizing transferable knowledge learned from a label-scarce source graph. However, most existing studies primarily concentrate on aligning feature distributions directly to extract domain-invariant features, while ignoring the utilization of the intrinsic structure information in graphs. Inspired by the significance of data structure information in enhancing models' generalization performance, this paper aims to investigate how to leverage the structure information to assist graph transfer learning. To this end, we propose an innovative framework called TFGDA. Specially, TFGDA employs a structure alignment strategy named STSA to encode graphs' topological structure information into the latent space, greatly facilitating the learning of transferable features. To achieve a stable alignment of feature distributions, we also introduce a SDA strategy to mitigate domain discrepancy on the sphere. Moreover, to address the overfitting issue caused by label scarcity, a simple but effective RNC strategy is devised to guide the discriminative clustering of unlabeled nodes. Experiments on various benchmarks demonstrate the superiority of TFGDA over SOTA methods. © 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

Research on WNN Greenhouse Temperature Prediction Method Based on GA

引用

Phyton-International Journal of Experimental Botany 2022年第10期91卷 2283-2296页

作者： Wenbin Dai Lina Wang Binrui Wang Xiaohong Cui Xue Li College of Mechanical and Electronic Engineering China Jiliang UniversityHangzhou310018China Key Laboratory of Intelligent Manufacturing Quality Big Data Tracing and Analysis of Zhejiang Province China Jiliang UniversityHangzhou310018China

Temperature in agricultural production has a direct impact on the growth of *** emergence of greenhouses has improved the impact of the original unpredictable changes in temperature,but the temperature modeling of greenhouses is still the main direction at *** network modeling relies on sufficient actual data to model greenhouses,but there is a widening gap in the application of different neural *** paper proposes a greenhouse temperature prediction model based on wavelet neural network with genetic algorithm(GA-WNN).With the simple network structure and the nonlinear adaptability of the wavelet basis function,wavelet neural network(WNN)improved model training speed and accuracy of prediction results compared with back propagation neural networks(BPNN),which was conducive to the prediction and control of short-term greenhouse temperature *** the same time,the genetic algorithm(GA)was introduced to globally optimize the initial weights of the original model,which improved the insensitivity of the model to the initial weights and thresholds,and improved the training speed and stability of the ***,simulation results for the greenhouse showed that the model training speed,prediction results accuracy and model stability of the GA-WNN in the greenhouse were improved in comparison to results obtained by the WNN and BPNN in the greenhouse.

关键词： Greenhouse temperature greenhouse modeling wavelet neural network genetic algorithm

来源：评论

学校读者我要写书评

暂无评论

A Model robustness optimization method based on adversarial sample detection 5

A Model robustness optimization method based on adversarial ...

引用

5th International Conference on Artificial Intelligence and Pattern Recognition, AIPR 2022

作者： Sun, Jiaze Long, Siyuan Ma, Xianyan Tang, Yanmei Xi'an University of Posts and Telecommunications Shaanxi Provincial Key Laboratory of Network Data Analysis and Intelligent Processing Xi'an Key Laboratory of Big Data and Intelligent Computing Xi'an710121 China Xi'an University of Posts and Telecommunications Xi'an710121 China

ISBN: (纸本)9781450396899

Deep neural networks are extremely vulnerable due to the existence of adversarial samples. It is a challenging problem to optimize the robustness of the model to protect deep neural networks from the threat of adversarial samples. To improve the model robustness, an integrated detection model of adversarial samples is designed to detect the existence of adversarial samples, which consists of a multi-classification detector and five single-classification detectors to perform double-layer detection of adversarial samples and intercept the adversarial samples finally sent to the image classification model. The detection experiments were conducted on the CIFAR-10 dataset for the adversarial samples generated by five attack algorithms: FGSM, BIM, DeepFool, JSMA, and C&W, and the detection success rate for all types of adversarial samples reached over 98.96%. After that, secondary attack experiments were conducted on the model, and the detection success rate of the model for the second attack adversarial samples reached over 92.36%. It provides an efficient and robust optimization method for deep neural network models in adversarial environments. © 2022 ACM.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

GoPIM: GCN-Oriented Pipeline Optimization for PIM Accelerators

GoPIM: GCN-Oriented Pipeline Optimization for PIM Accelerato...

引用

IEEE Symposium on High-Performance Computer Architecture

作者： Siling Yang Shuibing He Wenjiong Wang Yanlong Yin Tong Wu Weijian Chen Xuechen Zhang Xian-He Sun Dan Feng The State Key Laboratory of Blockchain and Data Security Zhejiang University Zhejiang Lab Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security Zhejiang Key Laboratory of Big Data Intelligent Computing Washington State University Vancouver Illinois Institute of Technology Huazhong University of Science and Technology Wuhan National Laboratory for Optoelectronics

ISBN: (数字)9798331506476

ISBN: (纸本)9798331506483

Graph convolutional networks (GCNs) are popular for a variety of graph learning tasks. ReRAM-based processing-in-memory (PIM) accelerators are promising to expedite GCN training owing to their in-situ computing capability. However, existing accelerators can be severely underutilized even with pipelines, due to the oversight of the skewed execution times of various GCN stages and the ignorance of skewed degrees of graph vertices. In this work, we propose GOPIM, a GCN-oriented pipeline optimization for PIM accelerators to expedite GCN training. First, GOPIM proposes an ML-based scheme that allocates crossbar resources to the most needed stages to streamline the overall pipeline. Second, GOPIM utilizes a selective vertex updating technique that evenly distributes vertices on crossbars by interleaved mapping. These techniques collectively reduce the overall execution time without losing much accuracy. We also provide a practical architecture design for GOPIM. Our experimental results show that, GoPIM achieves up to 191 × speedup and 16.1 × energy saving, compared to the state-of-the-art work.

关键词： Training Accuracy Graph convolutional networks Pipelines Energy conservation Computer architecture Machine learning Parallel processing Optimization

来源：评论

学校读者我要写书评

暂无评论

Research on key technologies of edge cache in virtual data space across WAN

引用

Frontiers of Computer Science 2023年第1期17卷 15-34页

作者： Jiantong HUO Yaowen XU Zhisheng HUO Limin XIAO Zhenxue HE State Key Laboratory of Software Development Environment Beihang UniversityBeijing 100191China School of Computer Science and Engineering Beihang UniversityBeijing 100191China High Performance Computing Center Beihang UniversityBeijing 100191China College of Software Beihang UniversityBeijing 100191China Hebei Key Laboratory of Agricultural Big Data Hebei Agricultural UniversityBaoding 071001China College of Computer Science and Technology Zhejiang UniversityHangZhou 310013China

The authors of this paper have previously proposed the global virtual data space system (GVDS) to aggregate the scattered and autonomous storage resources in China’s national supercomputer grid (National Supercomputing Center in Guangzhou, National Supercomputing Center in Jinan, National Supercomputing Center in Changsha, Shanghai Supercomputing Center, and Computer Network Information Center in Chinese Academy of Sciences) into a storage system that spans the wide area network (WAN), which realizes the unified management of global storage resources in China. At present, the GVDS has been successfully deployed in the China National Grid environment. However, when accessing and sharing remote data in the WAN, the GVDS will cause redundant transmission of data and waste a lot of network bandwidth resources. In this paper, we propose an edge cache system as a supplementary system of the GVDS to improve the performance of upper-level applications accessing and sharing remote data. Specifically, we first designs the architecture of the edge cache system, and then study the key technologies of this architecture: the edge cache index mechanism based on double-layer hashing, the edge cache replacement strategy based on the GDSF algorithm, the request routing based on consistent hashing method, and the cluster member maintenance method based on the SWIM protocol. The experimental results show that the edge cache system has successfully implemented the relevant operation functions (read, write, deletion, modification, etc.) and is compatible with the POSIX interface in terms of function. Further, it can greatly reduce the amount of data transmission and increase the data access bandwidth when the accessed file is located at the edge cache system in terms of performance, i.e., its performance is close to the performance of the network file system in the local area network (LAN).

关键词： virtual data space system wide area network edge cache redundant data transmission

来源：评论

学校读者我要写书评

暂无评论

IMPACT: Importance-Informed Prefetching and Caching for I/O-Bound DNN Training

引用

IEEE Transactions on Computers 2025年

作者： Chen, Weijian He, Shuibing Zhang, Ruidong Zhang, Xuechen Chen, Ping Yang, Siling Qu, Haoyang Zhan, Xuan Zhejiang University State Key Laboratory of Blockchain and Data Security China Institute of Blockchain and Data Security Hangzhou China Zhejiang Key Laboratory of Big Data Intelligent Computing China Washington State University Vancouver School of Engineering and Computer Science VancouverWA98686 United States

Fetching large amounts of DNN training data from storage systems causes high I/O latency and GPU stalls. Importance sampling can reduce data processing on GPUs while maintaining model accuracy, but current frameworks lack a prefetching and caching layer to optimize data fetches and cache management based on sample importance. This leads to unnecessary fetches, poor cache hit ratios, and random I/Os. We present ImPACT, an importance-informed prefetching and caching system, to accelerate I/O-bound DNN training. First, we propose an importance-informed prefetching technique to reduce the prefetching of unimportant data. Then, we introduce an importance-aware caching layer, partitioned into two regions: H-cache and L-cache, which store samples of high importance and low importance respectively. Rather than using recency or frequency, we manage data items in H-cache according to their corresponding sample importance. When there is a cache miss in L-cache, we use sample substitutability and dynamic packaging to improve the cache hit ratio and reduce the number of random I/Os. Our experimental results show that ImPACT has a negligible impact on training accuracy while speeding up DNN training by up to 3.5× compared to state-of-the-art prefetching and caching systems. © 1968-2012 IEEE.

关键词： Cache memory

来源：评论

学校读者我要写书评

暂无评论

How Do LLMs Acquire New Knowledge? A Knowledge Circuits Perspective on Continual Pre-Training

arXiv

引用

arXiv 2025年

作者： Ou, Yixin Yao, Yunzhi Zhang, Ningyu Jin, Hui Sun, Jiacheng Deng, Shumin Li, Zhenguo Chen, Huajun Zhejiang University China Huawei Noah’s Ark Lab Canada National University of Singapore NUS-NCS Joint Lab Singapore Zhejiang Key Laboratory of Big Data Intelligent Computing China

Despite exceptional capabilities in knowledge-intensive tasks, Large Language Models (LLMs) face a critical gap in understanding how they internalize new knowledge, particularly how to structurally embed acquired knowledge in their neural computations. We address this issue through the lens of knowledge circuit evolution, identifying computational subgraphs that facilitate knowledge storage and processing. Our systematic analysis of circuit evolution throughout continual pre-training reveals several key findings: (1) the acquisition of new knowledge is influenced by its relevance to preexisting knowledge;(2) the evolution of knowledge circuits exhibits a distinct phase shift from formation to optimization;(3) the evolution of knowledge circuits follows a deep-to-shallow pattern. These insights not only advance our theoretical understanding of the mechanisms of new knowledge acquisition in LLMs, but also provide potential implications for improving continual pre-training strategies to enhance model performance Copyright © 2025, The Authors. All rights reserved.

关键词： Knowledge acquisition

来源：评论

学校读者我要写书评

暂无评论

MLLM CAN SEE? DYNAMIC CORRECTION DECODING FOR HALLUCINATION MITIGATION

arXiv

引用

arXiv 2024年

作者： Wang, Chenxi Chen, Xiang Zhang, Ningyu Tian, Bozhong Xu, Haoming Deng, Shumin Chen, Huajun Zhejiang University China Nanjing University of Aeronautics and Astronautics China National University of Singapore NUS-NCS Joint Lab Singapore Zhejiang Key Laboratory of Big Data Intelligent Computing China

Multimodal Large Language Models (MLLMs) frequently exhibit hallucination phenomena, but the underlying reasons remain poorly understood. In this paper, we present an empirical analysis and find that, although MLLMs incorrectly generate the objects in the final output, they are actually able to recognize visual objects in the preceding layers. We speculate that this may be due to the strong knowledge priors of the language model suppressing the visual information, leading to hallucinations. Motivated by this, we propose a novel dynamic correction decoding method for MLLMs (DeCo), which adaptively selects the appropriate preceding layers and proportionally integrates knowledge into the final layer to adjust the output logits. Note that DeCo is model agnostic and can be seamlessly incorporated with various classic decoding strategies and applied to different MLLMs. We evaluate DeCo on widely-used benchmarks, demonstrating that it can reduce hallucination rates by a large margin compared to baselines, highlighting its potential to mitigate hallucinations1 Copyright © 2024, The Authors. All rights reserved.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

State-of-Charge Estimation of Lithium Battery Based on Deep Residual Shrinkage Networks and a Variant Long Short Term Memory Neural Network

State-of-Charge Estimation of Lithium Battery Based on Deep ...

引用

第35届中国控制与决策会议

作者： Penghua Li Yihui Zhang Key Laboratory of Intelligent Computing for Big Data College of Automation Chongqing University of Posts and Telecommunications

Accurate state-of-charge(SOC) estimation,which is critical to ensuring the safe and reliable operation of battery management systems in electric vehicles,is still a challenging task due to sophisticated battery dynamics and ever-changing ambient *** paper proposes a method combining DRSN and ASTLSTM neural networks to estimate lithium batteries' state of charge(SOC).The proposed method can not only utilize the historical information of the data but also eliminate unimportant features by inserting soft thresholding into the deep *** reveal the performance of the proposed method,the results are compared with several other methods from the *** comparative experimental results show that the proposed method achieves a lower root mean square error(RMSE) of 0.83% with normal data and RMSE of 0.98% with noisy data,respectively.

关键词： Lithium Battery State of Charge Recurrent Neural Network Deep Residual Shrinking Network

来源：评论

学校读者我要写书评

暂无评论

A Self-decoupled Interpretable Prediction Framework for Highly-Variable Cloud Workloads 28th

A Self-decoupled Interpretable Prediction Framework for Hig...

引用

28th International Conference on database Systems for Advanced Applications, DASFAA 2023

作者： Wang, Bingchao Shi, Xiaoyu Shang, Mingsheng Chongqing Key Laboratory of Big Data and Intelligent Computing Chongqing Institute of Green and Intelligent Technology Chinese Academy of Sciences Chongqing400714 China Chongqing School University of Chinese Academy of Sciences Chongqing400714 China Chongqing Key Laboratory of Computational Intelligence Chongqing University of Posts and Telecommunications Chongqing400065 China

ISBN: (纸本)9783031306365

Cloud workloads prediction plays a crucial role in the various tasks of cloud computing, such as resource scheduling, performance optimization, cost management, etc. However, current time series prediction methods suffer instability and inefficiency issues when addressing cloud workloads, due to the high variability of workload patterns and the high fluctuation within a workload. To address these issues, we propose DeIP4CW, a Self-Decoupled Interpretable Prediction framework for highly-variable Cloud Workloads. It can accurately forecast future job arrival rates in a cloud environment. The core idea of DeIP4CW is to first introduce the periodic and residual states as hidden variables to decouple complicated dependencies in cloud workload signals. Then it adopts a deep expansion learning framework with the block structure to perform workload prediction layer by layer. Each block consists of some periodic modules and some compensation modules. The periodic module with a self-attention mechanism can effectively capture the global trend of cloud workload, while the compensation module is employed to compensate for the local volatility information. Moreover, our two customized modules also have interpretable abilities, such as attributing the predictions to either global trends or local compensation. We conduct extensive experiments on the real-world cloud workload traces to evaluate the effectiveness of the proposed DeIP4CW. The experimental results demonstrate the DeIP4CW achieves significant improvements over the best baseline in most cases, and the error reduction can even reach up to 20.66%. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Cloud computing

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：