检索结果-内蒙古大学图书馆

YOLOO: You Only Learn from Others Once

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Gu, Lipeng Wei, Mingqiang Yan, Xuefeng Zhu, Dingkun Zhao, Wei Xie, Haoran Liu, Yong-Jin School of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing China Shenzhen Institute of Research Nanjing University of Aeronautics and Astronautics Shenzhen China The School of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing China The Collaborative Innovation Center of Novel Software Technology and Industrialization Nanjing China The School of Computer Science Jiangsu University of Technology Changzhou China The Division of Artificial Intelligence School of Data Science Lingnan University New Territories Hong Kong The MOE-Key Laboratory of Pervasive Computing Department of Computer Science and Technology Tsinghua University Beijing China

Multi-modal 3D multi-object tracking (MOT) typically necessitates extensive computational costs of deep neural networks (DNNs) to extract multi-modal representations. In this paper, we propose an intriguing question: May we learn from multiple modalities only during training to avoid multi-modal input in the inference phase? To answer it, we propose YOLOO, a novel multi-modal 3D MOT paradigm: You Only Learn from Others Once. YOLOO empowers the point cloud encoder to learn a unified tri-modal representation (UTR) from point clouds and other modalities, such as images and textual cues, all at once. Leveraging this UTR, YOLOO achieves efficient tracking solely using the point cloud encoder without compromising its performance, fundamentally obviating the need for computationally intensive DNNs. Specifically, YOLOO includes two core components: a unified tri-modal encoder (UTEnc) and a flexible geometric constraint (F-GC) module. UTEnc integrates a point cloud encoder with image and text encoders adapted from pretrained CLIP. It seamlessly fuses point cloud information with rich visual-textual knowledge from CLIP into the point cloud encoder, yielding highly discriminative UTRs that facilitate the association between trajectories and detections. Additionally, FGC filters out mismatched associations with similar representations but significant positional discrepancies. It further enhances the robustness of UTRs without requiring any scene-specific tuning, addressing a key limitation of customized geometric constraints (e.g., 3D IoU). Lastly, high-quality 3D trajectories are generated by a traditional data association component. By integrating these advancements into a multi-modal 3D MOT scheme, our YOLOO achieves substantial gains in both robustness and efficiency. Comprehensive experiments on KITTI and Waymo demonstrate the superior performance of our YOLOO over its twenty competitors. Copyright © 2024, The Authors. All rights reserved.

关键词： Deep neural networks

Graphology based handwritten character analysis for human behaviour identification

学校读者我要写书评

暂无评论

CAAI Transactions on Intelligence technology 2020年第1期5卷 55-65页

作者： Subhankar Ghosh Palaiahnakote Shivakumara Prasun Roy Umapada Pal Tong Lu Indian Statistical Institute Kolkata 700108India Faculty of Computer Science and Information Technology University of Malaya50603 Kuala LumpurMalaysia National Key Lab for Novel Software Technology Nanjing University210093 NanjingPeople’s Republic of China

Graphology-based handwriting analysis to identify human behavior,irrespective of applications,is *** existing methods that use characters,words and sentences for behavioural analysis with human intervention,we propose an automatic method by analysing a few handwritten English lowercase characters from a to z to identify person *** proposed method extracts structural features,such as loops,slants,cursive,straight lines,stroke thickness,contour shapes,aspect ratio and other geometrical properties,from different zones of isolated character images to derive the hypothesis based on a dictionary of Graphological *** derived hypothesis has the ability to categorise the personal,positive,and negative social aspects of an *** evaluate the proposed method,an automatic system is developed which accepts characters from a to z written by different individuals across different genders and age *** automatic privacy projected system is available on the website(http://***).For quantitative evaluation of the proposed method,several people are requested to use the system to check their characteristics with the system automatic response based on his/her handwriting by choosing to agree or disagree *** automatic system receives 5300 responses from the users,for which,the proposed method achieves 86.70%accuracy.

关键词： behaviour character analysis

HYBRID VALUE ESTIMATION FOR OFF-POLICY EVALUATION AND OFFLINE REINFORCEMENT LEARNING

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Jin, Xue-Kun Liu, Xu-Hui Jiang, Shengyi Yu, Yang National Key Laboratory for Novel Software Technology Nanjing University Nanjing China Department of Computer Science The University of Hong Kong Hong Kong Polixir.ai

Value function estimation is an indispensable subroutine in reinforcement learning, which becomes more challenging in the offline setting. In this paper, we propose Hybrid Value Estimation (HVE) to reduce value estimation error, which trades off bias and variance by balancing between the value estimation from offline data and the learned model. Theoretical analysis discloses that HVE enjoys a better error bound than the direct methods. HVE can be leveraged in both off-policy evaluation and offline reinforcement learning settings. We, therefore, provide two concrete algorithms Off-policy HVE (OPHVE) and Model-based Offline HVE (MOHVE), respectively. Empirical evaluations on MuJoCo tasks corroborate the theoretical claim. OPHVE outperforms other off-policy evaluation methods in all three metrics measuring the estimation effectiveness, while MOHVE achieves better or comparable performance with state-of-the-art offline reinforcement learning algorithms. We hope that HVE could shed some light on further research on reinforcement learning from fixed data. © 2022, CC BY.

关键词： Reinforcement learning

BPMN++: Comprehensive Business Process Modeling for Industrial Internet Application

学校读者我要写书评

暂无评论

BPMN++: Comprehensive Business Process Modeling for Industri...

IEEE International Conference on Big Data and Cloud Computing (BdCloud)

作者： Hangyu Cheng Guosheng Kang Jianxun Liu Yiping Wen Buqing Cao Zhen Wang School of Computer Science and Engineering Hunan University of Science and Technology Xiangtan China Hunan Provincial Key Lab. for Services Computing and Novel Software Technology HNUST Xiangtan China

Business process management is the end-to-end business process modeling, analysis, and optimization to achieve business goals. Business Process Modeling Notation (BPMN) is usually used to represent business process models. However, BPMN cannot support the intuitive modeling for some typical tasks in complex business scenarios of the Industrial Internet context, increasing the communication cost between modelers and business personnel. The current solutions try to extend the attributes of BPMN elements with annotations, but a large number of annotations makes business process models complex and cumbersome. This paper proposes to extend new business process modeling elements for Industrial Internet application scenarios based on BPMN extension mechanism, generating an extended business process modeling language, namely BPMN++. And an extended process modeling tool is implemented, which is available online. A case study over a real-world Industrial Internet process validates the usefulness of this extension mechanism and the well understandability of business process models.

关键词： Industries Process modeling Annotations Internet Personnel Task analysis Optimization

Further Explanations on "SAT Requires Exhaustive Search"

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Dong, Qingxiu Zhou, Guangyan Xu, Ke School of Computer Science Peking University China Department of Mathematics and Statistics Beijing Technology and Business University China State Key Lab of Software Development Environment Beihang University China

Recently, Xu and Zhou [2023] introduced a constructive approach for exploring computational hardness, proving that SAT requires exhaustive search. In light of certain misinterpretations concerning the contributions and proofs in that paper, we focus on providing detailed explanations in this work. We begin by delineating the core innovation of the constructive approach, shedding light on the pivotal concept of algorithm designability. We address the overlooked white-box diagonalization method and highlight the concept of an almost independent solution space. In response to specific misunderstandings, such as the concerns surrounding the assumptions of Lemma 3.1, we offer comprehensive clarifications aimed at improving the comprehension of the proof. We are grateful for the feedback received on our prior paper and hope this work can foster a more well-informed discussion. Copyright © 2024, The Authors. All rights reserved.

关键词：

MACRO: Incentivizing Multi-Leader Game-Based Pareto-Efficient Crowdsourcing for Video Analytics

学校读者我要写书评

暂无评论

MACRO: Incentivizing Multi-Leader Game-Based Pareto-Efficien...

International Conference on Data Engineering

作者： Yu Chen Sheng Zhang Ziying Zhou Xiaokun Wang Yu Liang Ning Chen Yuting Yan Mingjun Xiao Jie Wu Zhuzhong Qian Harry Xu State Key Lab. for Novel Software Technology Nanjing University P.R. China School of Systems Science Beijing Normal University China School of Computer and Electronic Information / School of Artificial Intelligence Nanjing Normal University China School of Computer Science and Technology University of Science and Technology of China P.R. China Center for Networked Computing Temple University USA Computer Science Department Samueli School of Engineering University of California Los Angeles USA

ISBN: (数字)9798350317152

ISBN: (纸本)9798350317169

In recent years, many crowdsourcing platforms have emerged, using the resources of recruited workers to perform diverse outsourcing tasks, where the video analytics attracts much attention due to its practical implications. For maximum profits, platforms carefully choose the workers and determine the video analytics configurations to ensure accuracy; meanwhile, workers possess the flexibility to tailor the configurations for their indivi-dual gains, which makes it hard for platforms to optimize their profits considering the platform-worker conflicts. In this paper, we design an incentive mechanism for Multi-leader game-based video Analytics upon CROwdsourcing, named MACRO, to over-come the above situation. Under that mechanism, we first formu-late the utility optimization problems for platforms and workers, respectively. We then propose a dual ascent-based method to op-timally determine the video analytics configurations for a multi-platform game, ensuring Pareto efficiency. Moreover, in the context of a multi-leader game involving platform-worker conflicts, we design an incentive function with its incentive factor update strategy and propose an ADMM-based approach for maximizing incentives that motivate workers to contribute to the platforms' profits. Rigorous proofs demonstrate the linear convergence of the MACRO to the multi-leader Stackelberg equilibrium. Trace-driven experiments show that MACRO improves the Pareto efficiency by 26.3%, outperforming other approaches.

关键词： Crowdsourcing Accuracy Visual analytics Games Data engineering Outsourcing Task analysis

Migrant Resettlement by Evolutionary Multi-objective Optimization

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Liu, Dan-Xuan Gu, Yu-Ran Qian, Chao Mu, Xin Tang, Ke National Key Laboratory for Novel Software Technology Nanjing University Nanjing210023 China School of Artificial Intelligence Nanjing University Nanjing210023 China Peng Cheng Laboratory Shenzhen518000 China Shenzhen Key Laboratory of Computational Intelligence Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen518055 China

Migration has been a universal phenomenon, which brings opportunities as well as challenges for global development. As the number of migrants (e.g., refugees) increases rapidly in recent years, a key challenge faced by each country is the problem of migrant resettlement. This problem has attracted scientific research attention, from the perspective of maximizing the employment rate. Previous works mainly formulated migrant resettlement as an approximately submodular optimization problem subject to multiple matroid constraints and employed the greedy algorithm, whose performance, however, may be limited due to its greedy nature. In this paper, we propose a new framework MR-EMO based on Evolutionary Multi-objective Optimization, which reformulates Migrant Resettlement as a bi-objective optimization problem that maximizes the expected number of employed migrants and minimizes the number of dispatched migrants simultaneously, and employs a Multi-Objective Evolutionary Algorithm (MOEA) to solve the bi-objective problem. We implement MR-EMO using three MOEAs, the popular NSGA-II, MOEA/D as well as the theoretically grounded GSEMO. To further improve the performance of MR-EMO, we propose a specific MOEA, called GSEMO-SR, using matrix-swap mutation and repair mechanism, which has a better ability to search for feasible solutions. We prove that MR-EMO using either GSEMO or GSEMO-SR can achieve better theoretical guarantees than the previous greedy algorithm. Experimental results under the interview and coordination migration models clearly show the superiority of MR-EMO (with either NSGA-II, MOEA/D, GSEMO or GSEMO-SR) over previous algorithms, and that using GSEMO-SR leads to the best performance of MR-EMO. Copyright © 2023, The Authors. All rights reserved.

关键词： Optimization algorithms

Smoothed online convex optimization based on discounted-normal-predictor 22

学校读者我要写书评

暂无评论

Smoothed online convex optimization based on discounted-norm...

Proceedings of the 36th International Conference on Neural Information Processing Systems

作者： Lijun Zhang Wei Jiang Jinfeng Yi Tianbao Yang National Key Laboratory for Novel Software Technology Nanjing University Nanjing China and Peng Cheng Laboratory Shenzhen China National Key Laboratory for Novel Software Technology Nanjing University Nanjing China JD AI Research Beijing China Department of Computer Science and Engineering Texas A & M University College Station

ISBN: (纸本)9781713871088

In this paper, we investigate an online prediction strategy named as Discounted-Normal-Predictor [Kapralov and Panigrahy, 2010] for smoothed online convex optimization (SOCO), in which the learner needs to minimize not only the hitting cost but also the switching cost. In the setting of learning with expert advice, Daniely and Mansour [2019] demonstrate that Discounted-Normal-Predictor can be utilized to yield nearly optimal regret bounds over any interval, even in the presence of switching costs. Inspired by their results, we develop a simple algorithm for SOCO: Combining online gradient descent (OGD) with different step sizes sequentially by Discounted-Normal-Predictor. Despite its simplicity, we prove that it is able to minimize the adaptive regret with switching cost, i.e., attaining nearly optimal regret with switching cost on every interval. By exploiting the theoretical guarantee of OGD for dynamic regret, we further show that the proposed algorithm can minimize the dynamic regret with switching cost in every interval.

关键词：

Identity Based Encryption and Biometric Authentication Scheme for Secure Data Access in Cloud Computing

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2023年第2期21卷 254-259页

作者： Hongbing Cheng Chunming Rong Zhenghua Tan Qingkai Zeng State Key Laboratory for Novel Software Technology Nanjing University Nanjing China Department of Information Engineering Jiangsu City Vocation College Nanjing China Department of Electronic Engineering and Computer Science University of Stavanger Stavanger Norway Department of Electronic Systems Aalborg University Aalborg Denmark

Cloud computing will be a main information infrastructure in the future; it consists of many large datacenters which are usually geographically distributed and heterogeneous. How to design a secure data access for cloud computing platform is a big challenge. In this paper, we propose a secure data access scheme based on identity-based encryption and biometric authentication for cloud computing. Firstly, we describe the security concern of cloud computing and then propose an integrated data access scheme for cloud computing, the procedure of the proposed scheme include parameter setup, key distribution, feature template creation, cloud data processing and secure data access control. Finally, we compare the proposed scheme with other schemes through comprehensive analysis and simulation. The results show that the proposed data access scheme is feasible and secure for cloud computing.

关键词： Access control Cloud computing Identity-based encryption Biometrics (access control) Authentication Process control Memory