检索结果-内蒙古大学图书馆

ABCP: Automatic block-wise and channel-wise network pruning via joint search

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Li, Jiaqi Li, Haoran Chen, Yaran Ding, Zixiang Li, Nannan Ma, Mingjun Duan, Zicheng Zhao, Dongbing The State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China The School of Mechanical Engineering Beijing Institute of Technology Beijing100081 China The University of Chinese Academy of Sciences Beijing China The College of Engineering and Computer Science Australian National University ACT2601 Australia

Currently, an increasing number of model pruning methods are proposed to resolve the contradictions between the computer powers required by the deep learning models and the resource-constrained devices. However, most of the traditional rule-based network pruning methods can not reach a sufficient compression ratio with low accuracy loss and are time-consuming as well as laborious. In this paper, we propose Automatic Block-wise and Channel-wise Network Pruning (ABCP1) to jointly search the block-wise and channel-wise pruning action with deep reinforcement learning. A joint sample algorithm is proposed to simultaneously generate the pruning choice of each residual block and the channel pruning ratio of each convolutional layer from the discrete and continuous search space respectively. The best pruning action taking both the accuracy and the complexity of the model into account is obtained finally. Compared with the traditional rule-based pruning method, this pipeline saves human labor and achieves a higher compression ratio with lower accuracy loss. Tested on the mobile robot detection dataset, the pruned YOLOv3 model saves 99.5% FLOPs, reduces 99.5% parameters, and achieves 37.3× speed up with only 2.8% mAP loss. The results of the transfer task on the sim2real detection dataset also show that our pruned model has much better robustness performance. Copyright © 2021, The Authors. All rights reserved.

关键词： Reinforcement learning

Simulations and Experiments of Time-Varying Takagi Factorization via Zhang Neural Network

学校读者我要写书评

暂无评论

Simulations and Experiments of Time-Varying Takagi Factoriza...

IEEE Advanced Information Technology, Electronic and Automation control Conference (IAEAC)

作者： Zanyu Tang Tianyu Zeng Zhenyu Li Jianrong Chen Yunong Zhang School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Guangdong Key Laboratory of Modern Control Technology Guangzhou China College of Information Science and Engineering Jishou University Jishou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Guangzhou China School of Public Health Administration Youjiang Medical University for Nationalities Baise China

ISBN: (数字)9781728180281

ISBN: (纸本)9781728180298

The problem of Takagi factorization (TF) for complex time-varying symmetric matrix is proposed in this paper. To solve this difficult and challenging problem, a continuous-time TF (CTTF) model of derivative equation system form is proposed via Zhang neural network (ZNN) method. For the sake of digital hardware implementation, we also build up the corresponding discrete-time (DT) ZNN solution models with different precisions of the derivative equation system discretization. Specifically, an eight-instant DTTF (EIDTTF) model are proposed. We also comparatively use two other multi-instant (e.g., four-instant and six-instant, i.e., FIDTTF and SIDTTF, respectively) models by discretizing the CTTF model. The effective solution step-size intervals of the proposed DTTF models are further discussed lastly. The results of simulations and experiments substantiate the correctness, effectiveness and superiority of the proposed CTTF and DTTF models.

关键词： Symmetric matrices Conferences Neural networks Hardware Numerical models Mathematical model Information technology

Regret and cumulative constraint violation analysis for online convex optimization with long term constraints

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Yi, Xinlei Li, Xiuxian Yang, Tao Xie, Lihua Chai, Tianyou Johansson, Karl H. School of Electrical Engineering and Computer Science and Digital Futures KTH Royal Institute of Technology Stockholm Sweden Department of Control Science and Engineering College of Electronics and Information Engineering Shanghai Research Institute for Intelligent Autonomous Systems Institute for Advanced Study Tongji University Shanghai China State Key Laboratory of Synthetical Automation for Process Industries Northeastern University Shenyang China School of Electrical and Electronic Engineering Nanyang Technological University Singapore Singapore

This paper considers online convex optimization with long term constraints, where constraints can be violated in intermediate rounds, but need to be satisfied in the long run. The cumulative constraint violation is used as the metric to measure constraint violations, which excludes the situation that strictly feasible constraints can compensate the effects of violated constraints. A novel algorithm is first proposed and it achieves an O(Tmax{c,1-c}) bound for static regret and an O(T (1-c)/2) bound for cumulative constraint violation, where c ϵ (0, 1) is a user-defined trade-off parameter, and thus has improved performance compared with existing results. Both static regret and cumulative constraint violation bounds are reduced to O(log(T )) when the loss functions are strongly convex, which also improves existing results. In order to achieve the optimal regret with respect to any comparator sequence, another algorithm is then proposed and it achieves the optimal O( p T (1 + PT )) regret and an O(√T) cumulative constraint violation, where PT is the path-length of the comparator sequence. Finally, numerical simulations are provided to illustrate the effectiveness of the theoretical results. Copyright © 2021, The Authors. All rights reserved.

关键词： Comparators (optical)

MVM3Det: A novel method for multi-view monocular 3D detection

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Li, Haoran Duan, Zicheng Ma, Mingjun Chen, Yaran Li, Jiaqi Zhao, Dongbin The State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China The University of Chinese Academy of Sciences Beijing China The College of Engineering and Computer Science Australian National University ACT2601 Australia The School of Mechanical Engineering Beijing Institute of Technology Beijing100081 China

Monocular 3D object detection encounters occlusion problems in many application scenarios, such as traffic monitoring, pedestrian monitoring, etc., which leads to serious false negatives. Multi-view object detection effectively solves this problem by combining data from different perspectives. However, due to label confusion and feature confusion, the orientation estimation of multi-view 3D object detection is intractable, which is important for object tracking and intention prediction. In this paper, we propose a novel multi-view 3D object detection method named MVM3Det which simultaneously estimates the 3D position and orientation of the object according to the multi-view monocular information. The method consists of two parts: 1) Position proposal network, which integrates the features from different perspectives into consistent global features through feature orthogonal transformation to estimate the position. 2) Multi-branch orientation estimation network, which introduces feature perspective pooling to overcome the two confusion problems during the orientation estimation. In addition, we present the first dataset for multi-view 3D object detection named MVM3D. Comparing with State-Of-The-Art (SOTA) methods on our dataset and public dataset WildTrack, our method achieves very competitive results. © 2021, CC BY.

关键词： Object recognition

X-scPAE: An explainable deep learning model for embryonic lineage allocation prediction based on single-cell transcriptomics revealing key genes in embryonic cell development

学校读者我要写书评

暂无评论

computers in Biology and Medicine 2025年 188卷 109787-109787页

作者： Liao, Kai Yan, Bowei Ding, Ziyin Huang, Jian Fan, Xiaodan Wu, Shanshan Chen, Changshui Li, Haibo The Central Laboratory of Birth Defects Prevention and Control The Affiliated Women and Children's Hospital of Ningbo University Ningbo315021 China Ningbo Key Laboratory for the Prevention and Treatment of Embryogenic Diseases The Affiliated Women and Children's Hospital of Ningbo University Ningbo315021 China Ningbo Key Laboratory of Genomic Medicine and Birth Defects Prevention The Affiliated Women and Children's Hospital of Ningbo University Ningbo315021 China MOE Engineering Research Center of Gene Technology School of Life Sciences Fudan University Shanghai200433 China Institutes of Biomedical Sciences Fudan University Shanghai200433 China Center for Reproductive Medicine The Affiliated Women and Children's Hospital of Ningbo University Ningbo315021 China College of Computer Science Chongqing University Chongqing400044 China

In single-cell transcriptomics research, accurately predicting cell lineage allocation and identifying differences between lineages are crucial for understanding cell differentiation processes and reducing early pregnancy miscarriages in humans. This paper introduces an explainable PCA-based deep learning attention autoencoder model, X-scPAE (eXplained Single Cell PCA - Attention Auto Encoder), which is built on the Counterfactual Gradient Attribution (CGA) algorithm. The model is designed to predict lineage allocation in human and mouse single-cell transcriptomic data, while identifying and interpreting gene expression differences across lineages to extract key genes. It first reduces dimensionality using Principal Component Analysis (PCA) and ranks the importance of principal components. An autoencoder is then employed for feature extraction, integrating an attention mechanism to capture interactions between features. Finally, the Counterfactual Gradient Attribution algorithm calculates the importance of each feature. The model achieved an accuracy of 0.945 on the test set and 0.977 on the validation set, with other metrics such as F1-score, Precision, and Recall all reaching 0.94. It significantly outperformed both baseline algorithms (XGBoost, SVM, RF, and LR) and advanced approaches like F-Score-SVM, CV2-LR, scChrBin, and TripletCell. Notably, the explainability analysis uncovered key lineage predictor genes for both humans and mice and identified crucial genes distinguishing between developmental stages and lineages. A logistic regression model built using the extracted key genes still achieved an AUROC of 0.92, surpassing the performance of other feature extraction methods, including F-Score, CV2, PCA, random feature selection, and the interpretability method Shapley. Lastly, ablation studies demonstrated the effectiveness of each model component. © 2025 The Authors

关键词： Mammals

Barrier lyapunov function-based fixed-time FTC for high-order nonlinear systems with predefined tracking accuracy

学校读者我要写书评

暂无评论

Research Square

Research Square 2021年

作者： Wang, Xiaolin Xu, Jihui Lv, Maolong Zhang, Lei Zhao, Zilong Department of Equipment Management UAV Engineering College Air Force Engineering University Shaanxi Xi’an710051 China Delft Center for Systems and Control Delft University of Technology Mekelweg 2 Delft2628 CD Netherlands Faculty of Electrical Engineering Mathematics and Computer Science Delft University of Technology Van Mourik Broekmanweg 6 Delft2628 XE Netherlands

This article proposes a fixed-time adaptive fault-tolerant control methodology for a larger class of high-order nonlinear systems subject to full-state constraints and actuator faults. In contrast with the state-of-the-art results, the distinguishing feature of our control design consists in proposing a novel high-order tan-type barrier Lyapunov function (BLF) which ensures state variables to be some asymmetric time-varying compact sets under tan-type constraints and expands the application range of tan-type BLF (i.e., from low-order to high-order, from symmetric time-invariant to asymmetric time-varying). Apart from this, the proposed control design ensures the tracking errors converge to specified residual sets within fixed-time and makes the size of the convergence regions of tracking errors adjustable a priori by means of a new BLF-based tuning function and a projection operator. A variable-separable lemma is delicately embedded into the control design to extract the control terms in a linear-like fashion which not only overcomes the difficulty that virtual control signals appear in a non-affine manner, but also solves the problem of actuator faults. Comparative simulations results finally validate the effectiveness of the proposed scheme. © 2021, CC BY.

关键词： Lyapunov functions

Reinforcement learning control of constrained dynamic systems with uniformly ultimate boundedness stability guarantee

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Han, Minghao Tian, Yuan Zhang, Lixian Wang, Jun Pan, Wei Department of Control Science and Engineering Harbin Institute of Technology China Department of Cognitive Robotics Delft University of Technology Netherlands Department of Computer Science University College London United Kingdom

Reinforcement learning (RL) is promising for complicated stochastic nonlinear control problems. Without using a mathematical model, an optimal controller can be learned from data evaluated by certain performance criteria through trial-and-error. However, the data-based learning approach is notorious for not guaranteeing stability, which is the most fundamental property for any control system. In this paper, the classic Lyapunov’s method is explored to analyze the uniformly ultimate boundedness stability (UUB) solely based on data without using a mathematical model. It is further shown how RL with UUB guarantee can be applied to control dynamic systems with safety constraints. Based on the theoretical results, both off-policy and on-policy learning algorithms are proposed respectively. As a result, optimal controllers can be learned to guarantee UUB of the closed-loop system both at convergence and during learning. The proposed algorithms are evaluated on a series of robotic continuous control tasks with safety constraints. In comparison with the existing RL algorithms, the proposed method can achieve superior performance in terms of maintaining safety. As a qualitative evaluation of stability, our method shows impressive resilience even in the presence of external disturbances. © 2020, CC-BY.

关键词： Reinforcement learning

Deep Learning in Cancer Diagnosis and Prognosis Prediction: A Minireview on Challenges, Recent Trends, and Future Directions

学校读者我要写书评

暂无评论

Computational and Mathematical Methods in Medicine 2021年第1期2021卷 9025470页

作者： Tufail, Ahsan Bin Ma, Yong-Kui Kaabar, Mohammed K. A. Martínez, Francisco Junejo, A.R. Ullah, Inam Khan, Rahim School of Electronics and Information Engineering Harbin Institute of Technology Harbin150001 China Department of Electrical and Computer Engineering COMSATS University Islamabad Sahiwal Campus Sahiwal57000 Pakistan Gofa Camp Near Gofa Industrial College and German Adebabay Nifas Silk-Lafto Addis Ababa26649 Ethiopia Institute of Mathematical Sciences Faculty of Science University of Malaya Kuala Lumpur50603 Malaysia Department of Applied Mathematics and Statistics Technological University of Cartagena Cartagena30203 Spain School of Control Science and Control Engineering Harbin Institute of Technology Harbin150001 China Changzhou Campus 213022 China

Deep learning (DL) is a branch of machine learning and artificial intelligence that has been applied to many areas in different domains such as health care and drug design. Cancer prognosis estimates the ultimate fate of a cancer subject and provides survival estimation of the subjects. An accurate and timely diagnostic and prognostic decision will greatly benefit cancer subjects. DL has emerged as a technology of choice due to the availability of high computational resources. The main components in a standard computer-aided design (CAD) system are preprocessing, feature recognition, extraction and selection, categorization, and performance assessment. Reduction of costs associated with sequencing systems offers a myriad of opportunities for building precise models for cancer diagnosis and prognosis prediction. In this survey, we provided a summary of current works where DL has helped to determine the best models for the cancer diagnosis and prognosis prediction tasks. DL is a generic model requiring minimal data manipulations and achieves better results while working with enormous volumes of data. Aims are to scrutinize the influence of DL systems using histopathology images, present a summary of state-of-the-art DL methods, and give directions to future researchers to refine the existing methods. © 2021 Ahsan Bin Tufail et al.

关键词： Forecasting

Formal analysis of QUIC handshake protocol using ProVerif

学校读者我要写书评

暂无评论

Formal analysis of QUIC handshake protocol using ProVerif

IEEE International Conference on Cyber Security and Cloud Computing (CSCloud)

作者： Jingjing Zhang Lin Yang Xianming Gao Qiang Wang College of Command and Control Engineering Army Engineering University of PLA Nanjing China National Key Laboratory of Science and Technology on Information System Security Chinese Academy of Military Science Beijing China Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen China

ISBN: (数字)9781728165509

ISBN: (纸本)9781728165516

This paper presents a formal analysis of the QUIC protocol. This newly proposed protocol is currently under the process of standardization, and it is still unknown whether the proposed protocol provides the claimed security guarantees. To fill this gap, we model the QUIC handshake protocol in the applied pi calculus and perform a comprehensive formal analysis using ProVerif. To the best of our knowledge, this is the first thorough study on the formal analysis of QUIC handshake protocol.

关键词： Security protocol QUIC Formal verification ProVerif formal analysis Formal verification Handshake Protocol pi calculus