检索结果-内蒙古大学图书馆

High Performance Big Data and intelligent Systems (HPBD&IS)

作者： Ming Zhang Jian Xu Jinzhong He Hong Qin AnnLab Institute of Semiconductors Chinese Academy of Sciences Beijing China Beijing Key Laboratory of Semiconductor Neural Network Intelligent Sensing and Computing Technology Beijing China College of Materials Science and Opto-Electronic Technology & School of Integrated Circuits University of Chinese Academy of Sciences Beijing China

Deep convolutional neural networks (DCNs) have recently experienced rapid development in the direction of lightweight and edge deployment. However, accelerators for DCNs face challenges in balancing computational and data bandwidth, leading to inefficient computation and high hardware costs. Additionally, different network structures make it challenging to design and reconfigure accelerators flexibly. To address these issues, this paper proposes a parallel-serial channel accelerator system, which resolves the low utilization of multipliers caused by small channels and inadequate bandwidth of fully connected layers. The results demonstrate that the proposed accelerator in this study maintains high computational performance and efficiency on typical DCNs. When implemented on Xilinx VCU128 at 200 MHz, the peak computational performance reaches 204.5 GOPS, with an efficiency of 0.37 GOPS/DSP and a maximum utilization rate of computing array up to 99.63%, surpassing previous works.

关键词：

来源：评论

学校读者我要写书评

暂无评论

RLEPSO: Reinforcement learning based Ensemble particle swarm optimizer 4

RLEPSO: Reinforcement learning based Ensemble particle swarm...

引用

4th International Conference on Algorithms, computing and Artificial Intelligence, ACAI 2021

作者： Yin, Shiyuan Liu, Yi Gong, GuoLiang Lu, Huaxiang Li, Wenchang Institute of Semiconductors Chinese Academy of Sciences Beijing China Institute of Semiconductors Chinese Academy of Sciences University of Chinese Academy of Sciences China Academy of Sciences Brain Science and Intelligent Technology Excellence Innovation Center Beijing Key Laboratory of Semiconductor Neural Network Intelligent Sensing and Computing Technology Beijing China

ISBN: (纸本)9781450385053

Evolution is the driving force behind the evolution of biological intelligence. Learning is the driving force behind human civilization. The combination of evolution and learning can form an entire natural world. Now, reinforcement learning has shown significant effects in many places. However, Currently, researchers in the field of optimization algorithms mainly focus on evolution strategies. And there is very little research on learning. Inspired by these ideas, this paper proposes a new particle swarm optimization algorithm Reinforcement learning based Ensemble particle swarm optimizer (RLEPSO) that combines reinforcement learning. The algorithm uses reinforcement learning for pre-training in the design phase to automatically find a more effective combination of parameters for the algorithm to run better and Complete optimization tasks faster. Besides, this algorithm integrates two robust particle swarm variants. And it sets the weight parameters for different algorithms to better adapt to the solution requirements of a variety of different optimization problems, which significantly improves the robustness of the algorithm. RLEPSO makes a certain number of sub-swarms to increase the probability of finding the global optimum and increasing the diversity of particle swarms. This proposed RLEPSO is evaluated on an optimization test functions benchmark set (CEC2013) with 28 functions and compared with other eight particle swarm optimization variants, including three state-of-the-art optimization algorithms. The results show that RLEPSO has better performance and outperforms all compared algorithms. © 2021 Owner/Author.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Larger receptive Field and Context Information for Pose estimation:Larger Gaussian kernel 5

Larger receptive Field and Context Information for Pose esti...

引用

5th International Conference on High Performance Big Data and intelligent Systems, HDIS 2023

作者： Ma, Junxiao Yang, Kai Ke, Zunwang Wang, Gang Zhang, Yugui Cao, Fengcai Zhou, Qi Fei, Yang School of Information Beijing Forestry University Beijing100083 China CGNPC Uranium Industry Development Co. Ltd. Beijing100020 China School of Software Xinjiang University Urumqi830046 China School of Computing and Data Engineering NingboTech University Ningbo315100 China AnnLab Institute of Semiconductors Chinese Academy of Sciences Beijing100083 China Beijing Key Laboratory of Semiconductor Neural Network Intelligent Sensing and Computing Technology Beijing100083 China School of Integrated Circuits University of Chinese Academy of Sciences Beijing100049 China Co. Ltd Beijing100176 China School of Semiconductor Science and Technology South China Normal University GuangZhou510631 China

ISBN: (纸本)9798350394160

The field of pose estimation has a wide range of application prospects in various industries in the current era. With the continuous development of deep learning techniques, the effects in the field of human pose estimation are constantly being optimized. However, due to the influence of factors such as occlusion and human posture variability, it is easy to lead to inaccurate identification of keypoints and even keypoints’ missing and matching errors, there is still a lot of optimization space for the effect of pose estimation. In this paper, we propose an improvement to generate heatmaps in the training process: increase the radius of the Gaussian kernel and adjust the smoothness of Gaussian distribution to get a larger receptive field, encourage the model to use more context information in prediction, and control other parameters unchanged on the premise of using heatmap regression. Improve keypoint detection performance by optimizing this item. In the experiment, for the COCO-Keypoint data set, we selected different methods published in recent years, and after our improvement, we obtained better accuracy and performance than the original method, which proves the validity of our method. ©2023 IEEE.

关键词： Gaussian distribution

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Strategies for Addressing Anomalous Exposure in Image Processing: The FARDBUNet Approach 5

Deep Learning Strategies for Addressing Anomalous Exposure i...

引用

5th International Conference on High Performance Big Data and intelligent Systems, HDIS 2023

作者： Zhou, Qi Yang, Kai Ke, Zunwang Wang, Gang Zhang, Yugui Jia, Yizhen Cao, Fengcai Ma, Junxiao Liu, Changlin Zhang, Kaijie Wu, Min School of Semiconductor Science and Technology South China Normal University GuangZhou510631 China CGNPC Uranium Industry Development Co. Ltd. Beijing100020 China School of Software Xinjiang University Urumqi830046 China School of Computing and Data Engineering NingboTech University Ningbo315100 China AnnLab Institute of Semiconductors Chinese Academy of Sciences Beijing100083 China Beijing Key Laboratory of Semiconductor Neural Network Intelligent Sensing and Computing Technology Beijing100083 China School of Integrated Circuits University of Chinese Academy of Sciences Beijing100049 China Co. Ltd Beijing100176 China School of Information Beijing Forestry University Beijing100083 China

ISBN: (纸本)9798350394160

In real-world scenarios, capturing scenes with excessive dynamic range often leads to the partial loss of highlight or dark area information due to irradiance variations and limitations in the capture capabilities of imaging devices caused by their size and material constraints. This phenomenon gives rise to anomalous exposure problems and triggers the loss of crucial initial details, thereby becoming a significant obstacle to achieving high image quality. Despite numerous researchers’ integration of exposure compensation techniques into the imaging process, the anomalous exposure problem remains unsolved, hindering progress in image processing. Consequently, considerable research is valuable in implementing exposure calibration techniques on visual perception and multimedia devices. This dissertation proposes FARDBUNet, a novel end-to-end exposure calibration model designed to address the anomalous exposure problem. A key innovation of FARDBUNet is the incorporation of a fused attention mechanism, enabling the model to prioritize regions needing enhanced correction while accounting for the global exposure characteristics of the image. Additionally, to tackle the challenge of inadequate information availability in the localized areas due to exposure issues, we introduce an atrous residual dense block that efficiently captures local information by expanding the receptive field. Comprehensive tests validate the effectiveness of our suggested model, which performs on par with leading techniques on expansive exposure datasets. ©2023 IEEE.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Acq: Improving Generative Data-Free Quantization Via Attention Correction

SSRN

引用

SSRN 2023年

作者： Li, Jixing Guo, Xiaozhou Dai, Benzhe Gong, Guoliang Jin, Min Chen, Gang Mao, Wenyu Lu, Huaxiang Institute of Semiconductors Chinese Academy of Sciences Beijing100083 China University of Chinese Academy of Sciences Beijing100089 China Semiconductor Neural Network Intelligent Perception and Computing Technology Beijing Key Laboratory Beijing100083 China Materials and Optoelectronics Research Center University of Chinese Academy of Sciences Beijing100083 China College of Microelectronics University of Chinese Academy of Sciences Beijing China

Data-free quantization aims to achieve model quantization without accessing any authentic sample. It is significant in an application-oriented context involving data privacy. Converting noise vectors into synthetic samples through a generator is a popular data-free quantization method, which is called generative data-free quantization. However, there is a difference in attention between synthetic samples and authentic samples. This is always ignored and restricts the quantization performance. First, since synthetic samples of the same class are prone to have homogenous attention, the quantized network can only learn limited modes of attention. Second, synthetic samples in eval mode and training mode exhibit different attention. Hence, the batch-normalization statistics matching tends to be inaccurate. ACQ is proposed in this paper to fix the attention of synthetic samples. An attention center position-condition generator is established regarding the homogenization of intra-class attention. Restricted by the attention center matching loss, the attention center position is treated as the generator’s condition input to guide synthetic samples in obtaining diverse attention. Moreover, we design adversarial loss of paired synthetic samples under the same condition to prevent the generator from paying overmuch attention to the condition, which may result in mode collapse. To improve the attention similarity of synthetic samples in different network modes, we introduce a consistency penalty to guarantee accurate BN statistics matching. The experimental results demonstrate that ACQ effectively improves the attention problems of synthetic samples. Under various training settings, ACQ achieves the best quantization performance. For the 4-bit quantization of Resnet18 and Resnet50, ACQ reaches 67.55% and 72.23% accuracy, respectively. © 2023, The Authors. All rights reserved.

关键词： Data privacy

来源：评论

学校读者我要写书评

暂无评论

DoctorGPT: A Large Language Model with Chinese Medical Question-Answering Capabilities

DoctorGPT: A Large Language Model with Chinese Medical Quest...

引用

High Performance Big Data and intelligent Systems (HPBD&IS)

作者： Wenqiang Li Lina Yu Min Wu Jingyi Liu Meilan Hao Yanjie Li AnnLab Institute of Semiconductors Chinese Academy of Sciences Beijing China School of Electronic Electrical and Communication Engineering & School of Integrated Circuits University of Chinese Academy of Sciences Beijing China Beijing Key Laboratory of Semiconductor Neural Network Intelligent Sensing and Computing Technology Beijing China

Large Language Models (LLMs) have made incredible strides recently in understanding and reacting to user intents. However, these models typically excel in English and have not been specifically trained for medical applications, leading to suboptimal performance in responding to medical inquiries such as diagnostic queries and drug recommendations. In this paper, we propose DoctorGPT, a domain-specific large language model tailored for medical question-answering tasks. DoctorGPT leverages the open-source Baichuan2 as its foundational model, undergoes extensive pre-training on medical encyclopedic data to incorporate medical knowledge, and subsequently undergoes fine-tuning on a dataset consisting of two million medical instruction-dialogue pairs to enhance its question-answering capabilities. When compared to general-purpose large models, DoctorGPT demonstrates significant advantages in Chinese medical question-answerinz (O&A) tasks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ACQ: Improving Generative Data-free Quantization Via Attention Correction

arXiv

引用

arXiv 2023年

作者： Li, Jixing Guo, Xiaozhou Dai, Benzhe Gong, Guoliang Jin, Min Chen, Gang Mao, Wenyu Lu, Huaxiang Institute of Semiconductors Chinese Academy of Sciences Beijing100083 China University of Chinese Academy of Sciences Beijing100049 China Semiconductor Neural Network Intelligent Perception and Computing Technology Beijing Key Laboratory Beijing100083 China Materials and Optoelectronics Research Center University of Chinese Academy of Sciences Beijing100049 China College of Microelectronics University of Chinese Academy of Sciences Beijing100049 China

Data-free quantization aims to achieve model quantization without accessing any authentic sample. It is significant in an application-oriented context involving data privacy. Converting noise vectors into synthetic samples through a generator is a popular data-free quantization method, which is called generative data-free quantization. However, there is a difference in attention between synthetic samples and authentic samples. This is always ignored and restricts the quantization performance. First, since synthetic samples of the same class are prone to have homogenous attention, the quantized network can only learn limited modes of attention. Second, synthetic samples in eval mode and training mode exhibit different attention. Hence, the batch-normalization statistics matching tends to be inaccurate. ACQ is proposed in this paper to fix the attention of synthetic samples. An attention center position-condition generator is established regarding the homogenization of intra-class attention. Restricted by the attention center matching loss, the attention center position is treated as the generator's condition input to guide synthetic samples in obtaining diverse attention. Moreover, we design adversarial loss of paired synthetic samples under the same condition to prevent the generator from paying overmuch attention to the condition, which may result in mode collapse. To improve the attention similarity of synthetic samples in different network modes, we introduce a consistency penalty to guarantee accurate BN statistics matching. The experimental results demonstrate that ACQ effectively improves the attention problems of synthetic samples. Under various training settings, ACQ achieves the best quantization performance. For the 4-bit quantization of Resnet18 and Resnet50, ACQ reaches 67.55% and 72.23% accuracy, respectively. Copyright © 2023, The Authors. All rights reserved.

关键词： Data privacy

来源：评论

学校读者我要写书评

暂无评论

Rec-Symnet: Symbolic network-Based Rectifiable Learning Framework for Symbolic Regression

SSRN

引用

SSRN 2023年

作者： Liu, Jingyi Li, Weijun Yu, Lina Wu, Min Sun, Linjun Li, Wenqiang Li, Yanjie Institute of Semiconductors Chinese Academy of Sciences Beijing100083 China Center of Materials Science and Optoelectronics Engineering School of Integrated Circuits University of Chinese Academy of Sciences Beijing100049 China Beijing Key Laboratory of Semiconductor Neural Network Intelligent Sensing and Computing Technology Beijing100083 China

Symbolic regression (SR) can be utilized to unveil the underlying mathematical expressions that describe a given set of observed data. At present, SR can be categorized into two methods: learning-from-scratch and learning-with-experience. Compared to learning-from-scratch, learning-with-experience yields results that are comparable to those of several benchmarks and incurs significantly lower time costs for obtaining expressions. However, the learning-with-experience model performs poorly in terms of unseen data distributions and lacks a rectification tool, apart from constant optimization, which exhibits limited performance. In this study, we propose a Symbolic network-based Rectifiable Learning Framework (Rec-SymNet) that possesses the ability to correct errors. Rec-SymNet adopts Symbolic network (SymNet) to represent an expression, and the encoding of SymNet is designed to provide supervised information, with numerous self-generated expressions, to train a policy net (PolicyNet). The training of PolicyNet can offer prior knowledge to guide effective searches. Subsequently, the incorrectly predicted expressions are revised via a rectification mechanism. This rectification mechanism endows Rec-SymNet with broader applicability. Experimental results demonstrate that our proposed method achieves the highest averaged coefficient of determination on self-generated datasets when compared with other state-of-the-art methods and yields more accurate results in public datasets. © 2023, The Authors. All rights reserved.

关键词： Regression analysis

来源：评论

学校读者我要写书评

暂无评论

Fitting Curves with Fractional Implicit Polynomials: A PSO-Assisted Monomial Combination Optimization Framework

Fitting Curves with Fractional Implicit Polynomials: A PSO-A...

引用

Artificial Intelligence technology (ACAIT), Asian Conference on

作者： Yuerong Tong Lina Yu Weijun Li Jingyi Liu Luyang Hou Linjun Sun Min Wu AnnLab Institute of Semiconductors Chinese Academy of Sciences Beijing China School of Materials Science and Optoelectronic Technology & School of Integrated Circuits University of Chinese Academy of Sciences Beijing China Beijing Key Laboratory of Semiconductor Neural Network Intelligent Sensing and Computing Technology Beijing China Beijing University of Posts and Telecommunications Beijing China

ISBN: (数字)9798350359145

ISBN: (纸本)9798350359152

Implicit polynomial can efficiently represent the object contour for the curve fitting, and fractional implicit polynomial (FIP) is capable of describing complex objects at lower degree. However, both of IP and FIP based curve fitting methods suffer from the problem of insufficient fitting accuracy because of spurious zero sets. In this paper, we propose a particle swarm optimization (PSO)-assisted monomial combination optimization framework to fit curves based on FIP. In the proposed PSO-FIP framework, we adopt a monomial combination optimization strategy to remove redundant monomials and reduce spurious zero sets accordingly. And then, we solve the combinatorial explosion problem by using PSO in the monomial combination optimization. We further propose a geometric distance metric to serve as the fitness function in PSO which overcomes the trivial solution problem of algebraic distance. We conduct extensive experiments to validate the PSO-FIP framework and the results demonstrate that our approach can achieve higher accuracy with fewer coefficients in curve fitting.

关键词： Measurement Fitting Noise Interference Polynomials Surface fitting IP networks

来源：评论

学校读者我要写书评

暂无评论

A Fault Detection Method for Tree Contact Single-Phase-To-Ground Fault

SSRN

引用

SSRN 2023年

作者： Wu, Juping Xv, Huikai Jin, Min Gong, Guoliang Chen, Gang Mao, Wenyu Lu, Huaxiang Chen, Tianxiang Institute of Semiconductors CAS Beijing100083 China University of Chinese Academy of Sciences Beijing China Semiconductor Neural Network Intelligent Perception and Computing Technology Beijing Key Laboratory Beijing China Collage of Microelectronics University of Chinese Academy of Sciences Beijing China Materials and Optoelectronics Research Center University of Chinese Academy of Sciences Beijing China Tianjin University of Science and Technology Tianjin300457 China College of Nuclear Technology and Automation Engineering Chengdu University of Technology Chengdu610059 China

High impedance fault(HIF), especially tree contact single-phase-to-ground fault(TSF), often occurs in resonant grounding systems and are difficult to detect. Therefore, there is an urgent need for an effective TSF detection technology. In this paper, by analyzing the equivalent circuits of different impedance faults in resonant grounding system, and analyzing the transient zero-sequence components of underdamped state and overdamped state under different impedance, the conclusion is drawn: When a TSF occurs, the transient zero-sequence voltage will have a unique " Beat frequency " characteristic, and the problem of covering this phenomenon due to the imbalance of distribution network is effectively solved by fast independent component analysis(FastICA). Based on this conclusion, A fault detection method for TSF based on FastICA is proposed. Finally, the study conducted real experiments using two different impedance trees, bamboo and pine, to verify the proposed detection method. The results of the experiment found clear support for that the detection method is accurate and effective. © 2023, The Authors. All rights reserved.

关键词： Fault detection

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：