检索结果-内蒙古大学图书馆

International Joint Conference on Neural Networks (IJCNN)

作者： Chengpeng Hu Jiyuan Pei Jialin Liu Xin Yao Research Institute of Trustworthy Autonomous Systems (RITAS) Southern University of Science and Technology Shenzhen China Department of Computer Science and Engineering Guangdong Key Laboratory of Brain-inspired Intelligent Computation Southern University of Science and Technology Shenzhen China

Evolutionary algorithms have been used to evolve a population of actors to generate diverse experiences for training reinforcement learning agents, which helps to tackle the temporal credit assignment problem and improves the exploration efficiency. However, when adapting this approach to address constrained problems, balancing the trade-off between the reward and constraint violation is hard. In this paper, we propose a novel evolutionary constrained reinforcement learning (ECRL) algorithm, which adaptively balances the reward and constraint violation with stochastic ranking, and at the same time, restricts the policy's behaviour by maintaining a set of Lagrange relaxation coefficients with a constraint buffer. Extensive experiments on robotic control benchmarks show that our ECRL achieves outstanding performance compared to state-of-the-art algorithms. Ablation analysis shows the benefits of introducing stochastic ranking and constraint buffer.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Terrain Classification Based on Wheel-terrain Interaction Measurements using Aside-Wheel Camera

Terrain Classification Based on Wheel-terrain Interaction Me...

引用

IEEE International Conference on robotics and Biomimetics

作者： Feng Xue Longteng Hu Chen Yao Jin Wei Yunzhou Li Peichen Wang Zheng Zhu Zhenzhong Jia Department of Mechanical and Energy Engineering Shenzhen Key Laboratory of Biomimetic Robotics and Intelligent Systems Southern University of Science and Technology (SUSTech) Shenzhen China SUSTech Guangdong Provincial Key Laboratory of Human-Augmentation and Rehabilitation Robotics in Universities Shenzhen China Harbin of Institute of Technology (HIT) Harbin China

The scientific applications over a decade have demonstrated the robust practicality of wheeled rovers, which are capable of undertaking exploration tasks in various unstructured terrains. However, sinkage and slippage are extremely common on sandy terrains. Fortunately, these risks can be mitigated in a timely manner through the perception of terrain types. Thus, terrain classification technology has become a key solution to addressing these issues. In this paper, we propose a new terrain classification pipeline for a rover over sandy terrains. It exploits the aside-wheel sensor suite to measure sinkage, contact angle, slip ratio, and normal force. Due to the strong correlation between contact characteristics and terrain types form our experiments, we can infer terrain types based on wheel-terrain interaction parameters. We train various machine learning models and obtained classification results, which is evaluated using the record dataset form a singlewheel testbed. Through Experimental comparison, most models achieve high accuracy (over 90%), we analyze the reasons behind performance discrepancies among different models and recommend LGBM as the preferred approach. In the end, we transfer the models to use data collected from rover field tests and achieve a higher level of accuracy, which proves the effectiveness of our approach.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Brain-inspired Fully Hardware Hopfield Neural Network based on Memristive Arrays

A Brain-inspired Fully Hardware Hopfield Neural Network base...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Zilu Wang Xin Yao Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen China Guangdong Provincial Key Laboratory of Brain-inspired Intelligent Computation Shenzhen China Research Institute of Trustworthy Autonomous Systems Shenzhen China School of Computer Science University of Birmingham Birmingham UK

In this work, we propose a fully hardware implementation of Hopfield neural network (HNN) based on memristive arrays, which adopts a bottom-up brain-inspired hardware framework. Memristors are used as key computing units to design cell modules in the low layer to play roles similar to neurons and synapses, so as to perform information integration, filtering, and plasticity of weights in a simple circuit structure and in-memory computing way. Cell modules are cascaded by the mode of encoding and mapping based on HNN in a structured regular circuit way to construct functional modules with braininspired parallel analog computing capacity in middle layers. Different functional modules perform information interaction according to the system's requirements and goals, thus realizing the overall system in the top layer. Our proposed HNN system is then used to solve combinatorial optimization problems. Different from other similar work, our system starts from a memristor-based brain-inspired framework and is implemented fully in hardware. The experimental results show that our work can not only improve the convergence speed, but also can be conveniently used to solve problems of different scales because of its good scalability. In addition, with hardware overhead and power consumption analysis, our system has been shown to be very hardware-friendly. Our work represents an advance towards a memristor-based hardware system with brain-inspired structure and high performance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Creating Own AI Datasets from Different Language Sources Efficiently: Advantages, Disadvantages and Best Pracitces to Make Own Datasets 6

Creating Own AI Datasets from Different Language Sources Eff...

引用

6th IEEE International Conference and Workshop Obuda on Electrical and Power Engineering, CANDO-EPE 2023

作者： Molnar, Gyorgy Cserko, Jazsef Nagy, Eniko Balogh, Zoltan Óbuda University Kandó Kálmán Faculty of Electrical Engineering Budapest Hungary Széchenyi University János Csere Apáczai Faculty Győr Hungary John von Neumann University Gamf Faculty of Engineering and Computer Science Kecskemét Hungary Győr Hungary Institute of Cyberphysical Systems Óbuda University Antal Bejczy Center for Intelligent Robotics John von Neumann Faculty Budapest Hungary Constantine the Philosopher University in Nitra Faculty of Natural Sciences and Informatics Department of Informatics Nitra94901 Slovakia

ISBN: (纸本)9798350328752

Artificial Intelligence (AI) has become a ubiquitous technology that has the potential to revolutionize many industries. However, AI requires access to vast amounts of data to achieve its full potential. By understanding these data sources and the best practices for working with them, we can unlock the full potential of AI and create groundbreaking solutions that benefit society as a whole. Data sources can be utilized to train AI algorithms, including structured and unstructured data, but finding relevant datasets is challenging. The main problem is the international sources. Most of the datasets are available only in English or other common languages. However these are good sources, but translating them is expensive and time-consuming. This paper shows how to generate own datasets from open-source data with translation. © 2023 IEEE.

关键词： Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

DISCODVT: Generating Long Text with Discourse-Aware Discrete Variational Transformer

DISCODVT: Generating Long Text with Discourse-Aware Discrete...

引用

2021 Conference on Empirical Methods in Natural Language Processing, EMNLP 2021

作者： Ji, Haozhe Huang, Minlie Department of Computer Science and Technology Institute for Artificial Intelligence State Key Lab of Intelligent Technology and Systems Beijing National Research Center for Information Science and Technology Tsinghua University Beijing100084 China

ISBN: (纸本)9781955917094

Despite the recent advances in applying pretrained language models to generate high-quality texts, generating long passages that maintain long-range coherence is yet challenging for these models. In this paper, we propose DISCODVT, a discourse-aware discrete variational Transformer to tackle the incoherence issue. DISCODVT learns a discrete variable sequence that summarizes the global structure of the text and then applies it to guide the generation process at each decoding step. To further embed discourse-aware information into the discrete latent representations, we introduce an auxiliary objective to model the discourse relations within the text. We conduct extensive experiments on two open story generation datasets and demonstrate that the latent codes learn meaningful correspondence to the discourse structures that guide the model to generate long texts with better long-range coherence. © 2021 Association for Computational Linguistics

关键词：

来源：评论

学校读者我要写书评

暂无评论

FeCAM: Exploiting the Heterogeneity of Class Distributions in Exemplar-Free Continual Learning

arXiv

引用

arXiv 2023年

作者： Goswami, Dipam Liu, Yuyang Twardowski, Bartlomiej van de Weijer, Joost Department of Computer Science Universitat Autònoma de Barcelona Spain Computer Vision Center Barcelona Spain University of Chinese Academy of Sciences China State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences China IDEAS-NCBR

Exemplar-free class-incremental learning (CIL) poses several challenges since it prohibits the rehearsal of data from previous tasks and thus suffers from catastrophic forgetting. Recent approaches to incrementally learning the classifier by freezing the feature extractor after the first task have gained much attention. In this paper, we explore prototypical networks for CIL, which generate new class prototypes using the frozen feature extractor and classify the features based on the Euclidean distance to the prototypes. In an analysis of the feature distributions of classes, we show that classification based on Euclidean metrics is successful for jointly trained features. However, when learning from non-stationary data, we observe that the Euclidean metric is suboptimal and that feature distributions are heterogeneous. To address this challenge, we revisit the anisotropic Mahalanobis distance for CIL. In addition, we empirically show that modeling the feature covariance relations is better than previous attempts at sampling features from normal distributions and training a linear classifier. Unlike existing methods, our approach generalizes to both many- and few-shot CIL settings, as well as to domain-incremental settings. Interestingly, without updating the backbone network, our method obtains state-of-the-art results on several standard continual learning benchmarks. Code is available at https://***/dipamgoswami/FeCAM. © 2023, CC BY.

关键词： Normal distribution

来源：评论

学校读者我要写书评

暂无评论

Two-Stage Robust Optimization Under Decision Dependent Uncertainty

引用

IEEE/CAA Journal of Automatica Sinica 2022年第7期9卷 1295-1306页

作者： Yunfan Zhang Feng Liu Yifan Su Yue Chen Zhaojian Wang João P.S.Catalão State Key Laboratory of Power System and Generation Equipment the Department of Electrical EngineeringTsinghua UniversityBeijing 100084China Department of Mechanical and Automation Engineering the Chinese University of Hong KongHong Kong SARChina Ministry of Education Key Laboratory of System Control and Information Processing the Department of AutomationShanghai Jiao Tong Universityand also with Shanghai Engineering Research Center of Intelligent Control and ManagementShanghai 200240China Faculty of Engineering of the University of Porto and Institute for Systems and Computer Engineering Technology and Science(INESC TEC)Porto 4200-465Portugal IEEE

In the conventional robust optimization(RO)context,the uncertainty is regarded as residing in a predetermined and fixed uncertainty *** many applications,however,uncertainties are affected by decisions,making the current RO framework *** paper investigates a class of two-stage RO problems that involve decision-dependent *** introduce a class of polyhedral uncertainty sets whose right-hand-side vector has a dependency on the here-and-now decisions and seek to derive the exact optimal wait-and-see decisions for the second-stage problem.A novel iterative algorithm based on the Benders dual decomposition is proposed where advanced optimality cuts and feasibility cuts are designed to incorporate the uncertainty-decision *** computational tractability,robust feasibility and optimality,and convergence performance of the proposed algorithm are guaranteed with theoretical *** motivating application examples that feature the decision-dependent uncertainties are ***,the proposed solution methodology is verified by conducting case studies on the pre-disaster highway investment problem.

关键词： Benders decomposition decision-dependent uncertainty endogenous uncertainty robust optimization(RO)

来源：评论

学校读者我要写书评

暂无评论

Magnetically Driven Capsule Robot for Multi-Targeted Biopsy and Drug Delivery in Stomach

Magnetically Driven Capsule Robot for Multi-Targeted Biopsy ...

引用

IEEE International Conference on robotics and Biomimetics

作者： Zhihui Xu Tanyong Wei Zongze Li Ding Huang Shoubin Liu Chengzhi Hu Department of Mechanical and Energy Engineering Shenzhen Key Laboratory of Biomimetic Robotics and Intelligent Systems Southern University of Science and Technology Shenzhen China Guangdong Provincial Key Laboratory of Human-Augmentation and Rehabilitation Robotics in Universities Southern University of Science and Technology Shenzhen China School of Mechanical Engineering and Automation Harbin Institute of Technology Shenzhen China

Capsule robots have emerged as a promising solution for non-invasive and painless screening of gastrointestinal diseases, offering significant advantages in patient comfort and reduced risk of cross-infection. Despite these advantages, existing capsule robots fall short in their ability to conduct targeted biopsies and deliver drugs to specific gastric lesions. To address this limitation, this paper presents the design and development of a magnetically driven capsule robot that leverages visual feedback to achieve accurate targeting for biopsy and drug administration in gastric lesions. Internal magnets, responsive to an external gradient magnetic field, enable the controlled propulsion of the capsule robot. A kinetic model of the capsule robot is analyzed, allowing it to recognize lesion contours through the endoscopic camera. This recognition determines the rotation angle or small adjustments required for the capsule robot's attitude and positioning. Experimental results in a simplified in vitro stomach model showcase the capability of the capsule robot to achieve targeted biopsy and drug delivery. These results highlight the potential feasibility of the proposed capsule robot in enhancing diagnostic accuracy and treatment efficacy with broader applications within the realm of gastric healthcare.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Recurrent Sliding Mode-Based State Observer Design for a Subset of Nonlinear systems

Recurrent Sliding Mode-Based State Observer Design for a Sub...

引用

International Conference on Control, Automation, robotics and Vision (ICARCV)

作者： Gael P. Atheupe Bhagyashri T. Gurjar Bruno Monsuez Renault Group and IP Paris Guyancourt FRANCE Department of Aerospace Engineering Intelligent Systems & Control Lab Indian Institute of Technology Bombay Mumbai India Computer science & systems engineering Lab (U2IS) of ENSTA Paris Institute Poly-technique de Paris (IP Paris) Palaiseau FRANCE

ISBN: (数字)9798331518493

ISBN: (纸本)9798331518509

This paper introduces a novel Sliding Mode-Based Observer tailored for a specific subset of nonlinear systems of order $\boldsymbol n$ featuring Lipschitz nonlinearities. The study establishes stability conditions that ensure convergence of the estimation error (s) in finite time until order $\boldsymbol n$ , thus, providing an accurate state (s) estimation without the necessity for disturbance matching conditions. Furthermore, the study presents an extension of the scope of application of the proposed method to tackle a unique scenario characterized by a time-varying and non-invertible function of the output dynamics of the system model. The effectiveness of the proposed observer is showcased through simulation examples.

关键词： Estimation error Automation Accuracy Observers Stability analysis Nonlinear dynamical systems Time-varying systems Robots Convergence

来源：评论

学校读者我要写书评

暂无评论

R2 Indicator and Deep Reinforcement Learning Enhanced Adaptive Multi-Objective Evolutionary Algorithm

arXiv

引用

arXiv 2024年

作者： Tahernezhad-Javazm, Farajollah Rankin, Debbie Bois, Naomi Du Smith, Alice E. Coyle, Damien Intelligent Systems Research Center School of Computing Engineering and Intelligent Systems Ulster University LondonderryBT48 7JL United Kingdom The Bath Institute for Augmented Human University of Bath BathBA2 7AY United Kingdom The Department of Industrial and Systems Engineering Department of Computer Science and Software Engineering Auburn University AuburnAL36849 United States

Choosing an appropriate optimization algorithm is essential to achieving success in optimization challenges. Here we present a new evolutionary algorithm structure that utilizes a reinforcement learning-based agent aimed at addressing these issues. The agent employs a double deep q-network to choose a specific evolutionary operator based on feedback it receives from the environment during optimization. The algorithm’s structure contains five single-objective evolutionary algorithm operators. This single-objective structure is transformed into a multi-objective one using the R2 indicator. This indicator serves two purposes within our structure: first, it renders the algorithm multi-objective, and second, provides a means to evaluate each algorithm’s performance in each generation to facilitate constructing the reinforcement learning-based reward function. The proposed R2-reinforcement learning multi-objective evolutionary algorithm (R2-RLMOEA) is compared with six other multiobjective algorithms that are based on R2 indicators. These six algorithms include the operators used in R2-RLMOEA as well as an R2 indicator-based algorithm that randomly selects operators during optimization. We benchmark performance using the CEC09 functions, with performance measured by inverted generational distance and spacing. The R2-RLMOEA algorithm outperforms all other algorithms with strong statistical significance (p © 2024, CC BY.

关键词： Reinforcement learning

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：