检索结果-内蒙古大学图书馆

FPGA-Based Multi-precision Architecture for Accelerating Large-Scale Floating-Point Matrix computing 17th

学校读者我要写书评

暂无评论

FPGA-Based Multi-precision Architecture for Accelerating Lar...

17th IFIP WG 10.3 International Conference on Network and Parallel computing, NPC 2020

作者： Zhang, Longlong Peng, Yuanxi Hu, Xiao Huang, Ahui Tian, Tian State Key Laboratory of High Performance Computing School of Computer National University of Defense Technology Changsha China Institute of Microelectronics School of Computer National University of Defense Technology Changsha China

ISBN: (纸本)9783030794774

Matrix computing plays a vital role in many scientific and engineering applications, but previous work can only handle the data with specified precision based on FPGA. This study first presents algorithms, data flows, and mapping strategies to match the hardware structure for matrix computing of different precisions. Then, we propose a unified multi-precision matrix computing unit core that can handle three precisions and three matrix operation modes and can be used as a coprocessor for large-scale matrix computing which has advantages of low storage and high efficiency. Finally, we build a complete matrix computing acceleration system and deploy it on FPGA using 128 processing elements (PEs). The experimental results show that the accelerator achieves a maximum frequency of 180 MHz, and matrix computing of double-precision, single-precision, and half-precision floating-point data performs 46.1 GFLOPS, 92.1 GFLOPS, and 184.3 GFLOPS respectively, which is superior to other current designs in terms of application range and performance. © 2021, IFIP International Federation for Information Processing.

关键词： Field programmable gate arrays (FPGA)

Personalized Federated Learning on Heterogeneous and Long-Tailed Data via Expert Collaborative Learning

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Lv, Fengling Shang, Xinyi Zhou, Yang Zhang, Yiqun Li, Mengke Lu, Yang Key Laboratory of Multimedia Trusted Perception and Efficient Computing Xiamen University China University College London United Kingdom A*STAR Institute of High Performance Computing Singapore School of Computer Science and Technology Guangdong University of Technology Guangzhou China Guangdong Laboratory of Artificial Intelligence and Digital Economy Shenzhen China Fujian Key Laboratory of Sensing and Computing for Smart City School of Informatics Xiamen University China

Personalized Federated Learning (PFL) aims to acquire customized models for each client without disclosing raw data by leveraging the collective knowledge of distributed clients. However, the data collected in real-world scenarios is likely to follow a long-tailed distribution. For example, in the medical domain, it is more common for the number of general health notes to be much larger than those specifically related to certain diseases. The presence of long-tailed data can significantly degrade the performance of PFL models. Additionally, due to the diverse environments in which each client operates, data heterogeneity is also a classic challenge in federated learning. In this paper, we explore the joint problem of global long-tailed distribution and data heterogeneity in PFL and propose a method called Expert Collaborative Learning (ECL) to tackle this problem. Specifically, each client has multiple experts, and each expert has a different training subset, which ensures that each class, especially the minority classes, receives sufficient training. Multiple experts collaborate synergistically to produce the final prediction output. Without special bells and whistles, the vanilla ECL outperforms other state-of-the-art PFL methods on several benchmark datasets under different degrees of data heterogeneity and long-tailed distribution. Copyright © 2024, The Authors. All rights reserved.

关键词： Federated learning

A Multi-view Impartial Decision Network for Frontotemporal Dementia Diagnosis

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Deng, Guoyao Zou, Ke Wang, Meng Yuan, Xuedong Ying, Sancong Fu, Huazhu National Key Laboratory of Fundamental Science on Synthetic Vision Sichuan University Sichuan China College of Computer Science Sichuan University Sichuan China Institute of High Performance Computing A*STAR Singapore

Frontotemporal Dementia (FTD) diagnosis has been successfully progress using deep learning techniques. However, current FTD identification methods suffer from two limitations. Firstly, they do not exploit the potential of multi-view functional magnetic resonance imaging (fMRI) for classifying FTD. Secondly, they do not consider the reliability of the multi-view FTD diagnosis. To address these limitations, we propose a reliable multi-view impartial decision network (MID-Net) for FTD diagnosis in fMRI. Our MID-Net provides confidence for each view and generates a reliable prediction without any conflict. To achieve this, we employ multiple expert models to extract evidence from the abundant neural network information contained in fMRI images. We then introduce the Dirichlet Distribution to characterize the expert class probability distribution from an evidence level. Additionally, a novel Impartial Decision Maker (IDer) is proposed to combine the different opinions inductively to arrive at an unbiased prediction without additional computation cost. Overall, our MID-Net dynamically integrates the decisions of different experts on FTD disease, especially when dealing with multi-view high-conflict cases. Extensive experiments on a high-quality FTD fMRI dataset demonstrate that our model outperforms previous methods and provides high uncertainty for hard-to-classify examples. We believe that our approach represents a significant step toward the deployment of reliable FTD decision-making under multi-expert conditions. We will release the codes for reproduction after acceptance. Copyright © 2023, The Authors. All rights reserved.

关键词： Diagnosis

SAM-U: Multi-box prompts triggered uncertainty estimation for reliable SAM in medical image

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Deng, Guoyao Zou, Ke Ren, Kai Wang, Meng Yuan, Xuedong Ying, Sancong Fu, Huazhu National Key Laboratory of Fundamental Science on Synthetic Vision Sichuan University Sichuan China College of Computer Science Sichuan University Sichuan China Institute of High Performance Computing A*STAR Singapore

Recently, Segmenting Anything Model has taken a significant step towards general artificial intelligence. Simultaneously, its reliability and fairness have garnered significant attention, particularly in the field of healthcare. In this study, we propose a multi-box prompt-triggered uncertainty estimation for SAM cues to demonstrate the reliability of segmented lesions or tissues. We estimate the distribution of SAM predictions using Monte Carlo with prior distribution parameters, employing different prompts as a formulation of test-time augmentation. Our experimental results demonstrate that multi-box prompts augmentation enhances SAM performance and provides uncertainty for each pixel. This presents a groundbreaking paradigm for a reliable SAM. Copyright © 2023, The Authors. All rights reserved.

关键词： Reliability

ANF: Attention-Based Noise Filtering Strategy for Unsupervised Few-Shot Classification 18th

学校读者我要写书评

暂无评论

ANF: Attention-Based Noise Filtering Strategy for Unsupervis...

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

作者： Ni, Guangsen Zhang, Hongguang Zhao, Jing Xu, Liyang Yang, Wenjing Lan, Long Institute for Quantum Information and State Key Laboratory of High Performance Computing College of Computer National University of Defense Technology Changsha410073 China Systems Engineering Institute AMS Beijing China

ISBN: (纸本)9783030893699

How to learn concepts from few-shot samples remains an open challenge in the deep learning era. The previous meta-learning methods require a large number of annotated samples in the training phase, which still contributes to high manual-labeling costs. In this paper, we propose a unsupervised few-shot learning framework and pointed out that negative queue constructed via randomly sampling contains many false-negative samples (noise), which has negative impacts on the model’s generalized performance especially when only few samples are available. Specially, we propose an Attention-based Noise Filtering (ANF) strategy to make momentum contrastive loss more applicable to few-shot learning scenario. In addition, we also propose a dynamic momentum update method, which can greatly improve the classification accuracy. Our evaluations demonstrate state-of-the-art unsupervised few-shot learning performance, which is comparable to supervised baseline models. © 2021, Springer Nature Switzerland AG.

关键词： computer vision

ADAPT: Adaptive distributed optimization approach for uploading data with redundancy in cooperative mobile cloud

学校读者我要写书评

暂无评论

ADAPT: Adaptive distributed optimization approach for upload...

作者： Wang, Ji Bao, Weidong Zhu, Xiaomin College of Systems Engineering National University of Defense Technology Changsha China State Key Laboratory of High Performance Computing National University of Defense Technology Changsha China

With the development of information technology and the ubiquity of mobile devices, increasing amounts of data are generated, processed, and transmitted by mobile devices. To alleviate the tension between the energy poverty of mobile devices and the increasing demand for transmitting data, the energy-efficient data transmission problem attracts considerable interests. Nonetheless, how to upload data with redundancy efficiently lacks a thorough study despite the wide existence of this problem in many situations like data storage among mobile devices and mobile crowd sensing. Since uploading redundant data brings little value while still consuming precious energy, it is important to design an efficient approach for mobile devices to upload data with redundancy cooperatively. In this work, we formulate the uploading data with redundancy in cooperative mobile cloud as an energy-constrained utility maximization problem. To solve this problem, we propose an adaptive distributed optimization approach consisting of the correlated upload decision and the online distributed scheduling algorithm. By the correlated upload decision, each mobile device can make adaptive decisions on how much data to upload and which data to upload according to its own observations independently. The online distributed scheduling algorithm enables mobile devices to optimally upload data. A series of simulation experiments are conducted to demonstrate the effectiveness of our approach. Finally, we test our approach on a real demo system to verify its practicability in reality. © 2019 John Wiley & Sons, Ltd.

关键词： Digital storage

Modeling Neural Networks Training Process with Markov Decision Process 2

学校读者我要写书评

暂无评论

Modeling Neural Networks Training Process with Markov Decisi...

2nd International Conference on Artificial Intelligence and computer Engineering, ICAICE 2021

作者： Bai, Yantao Liu, Wanwei Mao, Xinjun Liang, Zhen National University of Defense Technology Key Laboratory of Software Engineering for Complex Systems Changsha China College of Computer Science National University of Defense Technology Changsha China Institute for Quantum Information National University of Defense Technology State Key Laboratory of High Performance Computing Changsha China

ISBN: (纸本)9781665421867

With the development of computer technology, statistics-based machine learning method has made great break-throughs, and also improved the development of artificial intelligence. Nevertheless, as a very influential model, neural networks are still treated as 'black boxes'. The results of neural networks are extremely sensitive to the training samples, which lead to great challenges to the controllability of the algorithm. With the wide application of machine learning, demand for interpretability and controllability of neural networks algorithms is increasing. As a result, various scholars have tried to explain and verify neural networks algorithms based on formal methods in recent years. In this paper, a method (called MNNTP) is presented to model the training process of neural networks by using a Markov decision process (MDP). Through MNNTP, the neural networks are abstracted into the form of MDP, which makes notable contributions for verifying some mathematical properties of the neural networks. © 2021 IEEE.

关键词： Markov processes

Transformer doctor: diagnosing and treating vision transformers 24

学校读者我要写书评

暂无评论

Transformer doctor: diagnosing and treating vision transform...

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

作者： Jiacong Hu Hao Chen Kejia Chen Yang Gao Jingwen Ye Xingen Wang Mingli Song Zunlei Feng College of Computer Science and Technology Zhejiang University and State Key Laboratory of Blockchain and Data Security Zhejiang University College of Computer Science and Technology Zhejiang University School of Software Technology Zhejiang University Bangsheng Technology Co. Ltd. Electrical and Computer Engineering National University of Singapore College of Computer Science and Technology Zhejiang University and Bangsheng Technology Co. Ltd. College of Computer Science and Technology Zhejiang University and State Key Laboratory of Blockchain and Data Security Zhejiang University and Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security School of Software Technology Zhejiang University and State Key Laboratory of Blockchain and Data Security Zhejiang University and Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security

ISBN: (纸本)9798331314385

Due to its powerful representational capabilities, Transformers have gradually become the mainstream model in the field of machine vision. However, the vast and complex parameters of Transformers impede researchers from gaining a deep understanding of their internal mechanisms, especially error mechanisms. Existing methods for interpreting Transformers mainly focus on understanding them from the perspectives of the importance of input tokens or internal modules, as well as the formation and meaning of features. In contrast, inspired by research on information integration mechanisms and conjunctive errors in the biological visual system, this paper conducts an in-depth exploration of the internal error mechanisms of Transformers. We first propose an information integration hypothesis for Transformers in the machine vision domain and provide substantial experimental evidence to support this hypothesis. This includes the dynamic integration of information among tokens and the static integration of information within tokens in Transformers, as well as the presence of conjunctive errors therein. Addressing these errors, we further propose heuristic dynamic integration constraint methods and rule-based static integration constraint methods to rectify errors and ultimately improve model performance. The entire methodology framework is termed as Transformer Doctor, designed for diagnosing and treating internal errors within transformers. Through a plethora of quantitative and qualitative experiments, it has been demonstrated that Transformer Doctor can effectively address internal errors in transformers, thereby enhancing model performance. For more information, please visit https://***/.

关键词：

Verifying Safety of Neural Networks from Topological Perspectives

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Liang, Zhen Ren, Dejin Xue, Bai Wang, Ji Yang, Wenjing Liu, Wanwei National University of Defense Technology Institute for Quantum Information State Key Laboratory of High Performance Computing Hunan Changsha410073 China Chinese Academy of Sciences State Key Lab. of Computer Science BeijingBAI XUE China

Neural networks (NNs) are increasingly applied in safety-critical systems such as autonomous vehicles. However, they are fragile and are often ill-behaved. Consequently, their behaviors should undergo rigorous guarantees before deployment in practice. In this paper, we propose a set-boundary reachability method to investigate the safety verification problem of NNs from a topological perspective. Given an NN with an input set and a safe set, the safety verification problem is to determine whether all outputs of the NN resulting from the input set fall within the safe set. In our method, the homeomorphism property and the open map property of NNs are mainly exploited, which establish rigorous guarantees between the boundaries of the input set and the boundaries of the output set. The exploitation of these two properties facilitates reachability computations via extracting subsets of the input set rather than the entire input set, thus controlling the wrapping effect in reachability analysis and facilitating the reduction of computation burdens for safety verification. The homeomorphism property exists in some widely used NNs such as invertible residual networks (i-ResNets) and Neural ordinary differential equations (Neural ODEs), and the open map is a less strict property and easier to satisfy compared with the homeomorphism property. For NNs establishing either of these properties, our set-boundary reachability method only needs to perform reachability analysis on the boundary of the input set. Moreover, for NNs that do not feature these properties with respect to the input set, we explore subsets of the input set for establishing the local homeomorphism property and then abandon these subsets for reachability computations. Finally, some examples demonstrate the performance of the proposed *** Codes 68Q60, 68T07 Copyright © 2023, The Authors. All rights reserved.

关键词： Topology