检索结果-内蒙古大学图书馆

Learning Optimal "Pigovian Tax" in Sequential Social Dilemmas

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Hua, Yun Gao, Shang Li, Wenhao Jin, Bo Wang, Xiangfeng Zha, Hongyuan School of Computer Science and Technology East China Normal University Shanghai200092 China School of Data Science The Chinese University of Hong Kong Shenzhen China Shenzhen Institute of Artificial Intelligence and Robotics for Society Guangdong Shenzhen China School of Software Engineering Tongji University Shanghai201804 China School of Computer Science and Technology East China Normal University Shanghai200062 China

In multi-agent reinforcement learning, each agent acts to maximize its individual accumulated rewards. Nevertheless, individual accumulated rewards could not fully reflect how others perceive them, resulting in selfish behaviors that undermine global performance. The externality theory, defined as "the activities of one economic actor affect the activities of another in ways that are not reflected in market transactions," is applicable to analyze the social dilemmas in MARL. One of its most profound non-market solutions, "Pigovian Tax", which internalizes externalities by taxing those who create negative externalities and subsidizing those who create positive externalities, could aid in developing a mechanism to resolve MARL's social dilemmas. The purpose of this paper is to apply externality theory to analyze social dilemmas in MARL. To internalize the externalities in MARL, the Learning Optimal Pigovian Tax method (LOPT), is proposed, where an additional agent is introduced to learn the tax/allowance allocation policy so as to approximate the optimal "Pigovian Tax" which accurately reflects the externalities for all agents. Furthermore, a reward shaping mechanism based on the approximated optimal "Pigovian Tax" is applied to reduce the social cost of each agent and tries to alleviate the social dilemmas. Compared with existing state-of-the-art methods, the proposed LOPT leads to higher collective social welfare in both the Escape Room and the Cleanup environments, which shows the superiority of our method in solving social dilemmas. Copyright © 2023, The Authors. All rights reserved.

关键词： Taxation

Low-Crosstalk in 2D Waveguide Crossings by Metaheuristic Optimization Methods

学校读者我要写书评

暂无评论

Low-Crosstalk in 2D Waveguide Crossings by Metaheuristic Opt...

Optics and Photonics Conference (SBFoton IOPC), SBFoton International

作者： Luana da França Vieira C. E. Rubio-Mercedes Vitaly Felix Rodriguez Esquerre Department of Computer Science Federal University of Western Bahia Barreiras Brazil Physics Engineering Program State University of Mato Grosso do Sul Dourados Brazil Graduate School of Electrical Engineering Federal University of Bahia Salvador Brazil

ISBN: (数字)9798350388176

ISBN: (纸本)9798350388183

This work describes the investigation of planar waveguide crossings using metaheuristic methods associated with the two-dimensional finite element method (2D-FEM). An optimization and measurement process are performed on the power transmission and crosstalk of the waveguide crossings and then the values calculated are compared with other studies in the literature. The best performing adjusted crossing is optimally designed by applying the following metaheuristics: Particle Swarm Optimization (PSO), Grey Wolf Optimizer (GWO), and a hybrid algorithm combining the Greedy Randomized Adaptive Search Procedure and Simulated Annealing (GRASP-SA). Transmission efficiency results higher than 97% and crosstalk below the -50 dB over a wide wavelength range, with a footprint of $\mathbf{4.9}\times \mathbf{4.9}\ \boldsymbol{\mu} \mathbf{m}^{\mathbf{2}}$ .

关键词： Performance evaluation Power measurement Metaheuristics Crosstalk Power transmission Simulated annealing Silicon Planar waveguides Optical waveguides Photonics

Rlsnake: A hybrid reinforcement learning approach for road detection 24

学校读者我要写书评

暂无评论

Rlsnake: A hybrid reinforcement learning approach for road d...

2021 24th ISPRS Congress Commission III: Imaging Today, Foreseeing Tomorrow

作者： Botteghi, N. Sirmacek, B. Ünsalan, C. Robotics and Mechatronics Faculty of Electrical Engineering Mathematics and Computer Science University of Twente Netherlands Smart Cities School of Creative Technology Saxion University of Applied Sciences Netherlands Electrical and Electronics Engineering Faculty of Engineering Marmara University Turkey

Road network detection from very high resolution satellite and aerial images is highly important for diverse domains. Although an expert can label road pixels in a given image, this operation is prone to error and quite time consuming remembering that road maps must be updated regularly. Therefore, various computer vision based automated algorithms have been proposed in the last two decades. Nevertheless, due to the diversity of scenes, the field is still open for robust methods which might detect roads on different resolution images of different type of environments. In this study, we picked an earlier proposed road detection method which works based on traditional computer vision and probability theory algorithms. We improved it by further steps using reinforcement learning theory. With the help of the novel hybrid technique (traditional computer vision method combined with reinforcement learning based artificial intelligence), we achieved a solution that we call RLSnake. This new method can learn new image scenes and resolutions rapidly and can work reliably. We believe that the proposed RLSnake will be a significant step in the remote sensing field in order to develop solutions which might increase performance by combining the power of traditional and new techniques. © 2021 International Archives of the Photogrammetry, Remote Sensing and Spatial Information sciences - ISPRS Archives. All rights reserved.

关键词： Reinforcement learning

Effective Dimension Aware Fractional-Order Stochastic Gradient Descent for Convex Optimization Problems

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Partohaghighi, Mohammad Marcia, Roummel Chen, YangQuan Electrical Engineering and Computer Science graduate program University of California Merced United States Department of Applied Mathematics University of California Merced MercedCA United States Lab Department of Mechanical Engineering School of Engineering University of California MercedCA95343 United States

Fractional-order stochastic gradient descent (FOSGD) leverages a fractional exponent to capture long-memory effects in optimization, yet its practical impact is often constrained by the difficulty of tuning and stabilizing this exponent. In this work, we introduce 2SED Fractional-Order Stochastic Gradient Descent (2SEDFOSGD), a novel method that synergistically combines the Two-Scale Effective Dimension (2SED) algorithm with FOSGD to automatically calibrate the fractional exponent in a data-driven manner. By continuously gauging model sensitivity and effective dimensionality, 2SED dynamically adjusts the exponent to curb erratic oscillations and enhance convergence rates. Theoretically, we demonstrate how this dimension-aware adaptation retains the benefits of fractional memory while averting the sluggish or unstable behaviors frequently observed in naive fractional SGD. Empirical evaluations across multiple benchmarks confirm that our 2SED-driven fractional exponent approach not only converges faster but also achieves more robust final performance, suggesting broad applicability for fractional-order methodologies in large-scale machine learning and related domains. © 2025, CC BY.

关键词： Convex optimization

SoftTouch: A Sensor-Placement Framework for Soft Robotic Hands

学校读者我要写书评

暂无评论

SoftTouch: A Sensor-Placement Framework for Soft Robotic Han...

IEEE-RAS International Conference on Humanoid Robots

作者： Chao Li Nancy Pollard Department of Electrical and Computer Engineering Carnegie Mellon University Pittsburgh PA USA Robotics Institute School of Computer Science Carnegie Mellon University Pittsburgh PA USA

ISBN: (纸本)9798350309805

Sensor placement for grasping tasks in conventional robotic hands has been extensively studied, with goals including sensorizing essential contact areas or determining the effect of number of sensors on performance. However, with the new generation of dexterous soft robotic hands that deform to the shape of the object, the former frameworks may not be sufficient. In particular, we find that real-world experiments are essential to determine the value of different sensors and the effect of different sensor placements due to the complex interactions between the deformable robot body, sensor material properties, and sensor and task performance. In this paper, we propose a sensor-placement framework for dexterous soft robotic hands that is easily reconfigurable to different hand designs using off-the-shelf sensors. Our three-step framework selects and evaluates candidate sensor configurations to de-termine the effectiveness of sensors in each configuration for estimating qualitative and quantitative manipulation metrics. We tested our framework on a soft robotic hand to select the optimum sensor placement for a given set of manipulation patterns using force and inertial sensors. Our studies show that sensors placed at contact points are best for predicting the qualitative success of the manipulation. However, when it comes to estimating quantitative manipulation metrics, off-the-shelf sensors placed at contact points decrease performance for some manipulation types. This performance decrease may be due to the disturbance they create to surface texture, deformation patterns, and weight of soft robotic systems.

关键词： Measurement Sensor placement Shape Humanoid robots Soft robotics Robot sensing systems Surface texture

TS-CGNet: Temporal-Spatial Fusion Meets Centerline-Guided Diffusion for BEV Mapping

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Hong, Xinying Li, Siyu Zeng, Kang Shi, Hao Peng, Bomin Yang, Kailun Li, Zhiyong The College of Computer Science and Electronic Engineering Hunan University Changsha410082 China The School of Robotics The National Engineering Research Center of Robot Visual Perception and Control Technology Hunan University Changsha410082 China The State Key Laboratory of Extreme Photonics and Instrumentation Zhejiang University Hangzhou310027 China

Bird’s Eye View (BEV) perception technology is crucial for autonomous driving, as it generates top-down 2D maps for environment perception, navigation, and decision-making. Nevertheless, the majority of current BEV map generation studies focusing on visual map generation lack depth-aware reasoning capabilities. They exhibit limited efficacy in managing occlusions and handling complex environments, with a notable decline in perceptual performance under adverse weather conditions or low-light scenarios. Therefore, this paper proposes TS-CGNet, which leverages Temporal-Spatial fusion with Centerline-Guided diffusion. This visual framework, grounded in prior knowledge, is designed for integration into any existing network for building BEV maps. Specifically, this framework is decoupled into three parts: Local mapping system involves the initial generation of semantic maps using purely visual information;The Temporal-Spatial Aligner Module (TSAM) integrates historical information into mapping generation by applying transformation matrices;The Centerline-Guided Diffusion Model (CGDM) is a prediction module based on the diffusion model. CGDM incorporates centerline information through spatial-attention mechanisms to enhance semantic segmentation reconstruction. We construct BEV semantic segmentation maps by our methods on the public nuScenes and the robustness benchmarks under various corruptions. Our method improves 1.90%, 1.73%, and 2.87% for perceived ranges of 60×30m, 120×60m, and 240×60m in the task of BEV HD mapping. TS-CGNet attains an improvement of 1.92% for perceived ranges of 100×100m in the task of BEV semantic mapping. Moreover, TS-CGNet achieves an average improvement of 2.92% in detection accuracy under varying weather conditions and sensor interferences in the perception range of 240×60m. The source code will be publicly available at https://***/krabs-H/TS-CGNet. Copyright © 2025, The Authors. All rights reserved.

关键词： Decision making

A method to generate context information sets from analysis results with a unified abstraction model based on an extension of data enrichment scheme

学校读者我要写书评

暂无评论

A method to generate context information sets from analysis ...

作者： Park, Yoosang Mun, Jonghyeok Choi, Jongsun Choi, Jaeyoung Kim, Hoda School of Computer Science and Engineering Soongsil University Seoul Korea Republic of National Program of Excellence in Software center Chosun University Gwangju Korea Republic of

Context-awareness techniques generally deal with steps for processing information that can be gathered from perspectives of objects in real-world scenes in accordance with resolving heterogeneous devices and providing meaningful context information in recent computing domains such as Internet of Things, cloud computing, edge computing, and big data analysis. However, analysis applications are necessary to be improved in a way of cooperative approaches because they are independently executed far away from context-aware systems. This means that the stages are applied to process the data or the derived results, are isolated. That leads difficulties of management and integration for other software modules. As a result, a computing paradigm called edge computing has been suggested to compromise such isolations. For processing context information, it is necessary to establish knowledge parts with validation processes against the isolations. In this paper, we suggest a method for processing datasets in a view of building knowledge parts. The suggested method uses an extensible data enrichment scheme, which represents the relations for the object. In the experiment section, applications of the proposed method are shown in one of the healthcare community centers, and the data abstraction with the analysis jobs can be processed for each real-world object. © 2020 John Wiley & Sons, Ltd.

关键词： Knowledge representation

Iterative Shaping of Multi-Particle Aggregates based on Action Trees and VLM

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Lee, Hoi-Yin Zhou, Peng Duan, Anqing Yang, Chenguang Navarro-Alarcon, David Department of Mechanical Engineering The Hong Kong Polytechnic University KLN Hong Kong School of Advanced Engineering The Great Bay University Dongguan China Department of Robotics Mohamed Bin Zayed University of Artificial Intelligence United Arab Emirates Department of Computer Science University of Liverpool Liverpool United Kingdom

In this paper, we address the problem of manipulating multi-particle aggregates using a bimanual robotic system. Our approach enables the autonomous transport of dispersed particles through a series of shaping and pushing actions using robotically-controlled tools. Achieving this advanced manipulation capability presents two key challenges: high-level task planning and trajectory execution. For task planning, we leverage Vision Language Models (VLMs) to enable primitive actions such as tool affordance grasping and non-prehensile particle pushing. For trajectory execution, we represent the evolving particle aggregate's contour using truncated Fourier series, providing efficient parametrization of its closed shape. We adaptively compute trajectory waypoints based on group cohesion and the geometric centroid of the aggregate, accounting for its spatial distribution and collective motion. Through realworld experiments, we demonstrate the effectiveness of our methodology in actively shaping and manipulating multi-particle aggregates while maintaining high system cohesion. © 2025, CC BY-NC-SA.

关键词： Fourier series

InstaGraM: Instance-level Graph Modeling for Vectorized HD Map Learning

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Shin, Juyeb Jeong, Hyeonjun Rameau, Francois Kum, Dongsuk Robotics Program Korea Advanced Institute of Science and Technology Daejeon34051 Korea Republic of Graduate School of Mobility Korea Advanced Institute of Science and Technology Daejeon34051 Korea Republic of Computer Science department State University of New York - Korea Songdo21985 Korea Republic of

For scalable autonomous driving, a robust map-based localization system, independent of GPS, is fundamental. To achieve such map-based localization, online high-definition (HD) map construction plays a significant role in accurate estimation of the pose. Although recent advancements in online HD map construction have predominantly investigated on vectorized representation due to its effectiveness, they suffer from computational cost and fixed parametric model, which limit scalability. To alleviate these limitations, we propose a novel HD map learning framework that leverages graph modeling. This framework is designed to learn the construction of diverse geometric shapes, thereby enhancing the scalability of HD map construction. Our approach involves representing the map elements as an instance-level graph by decomposing them into vertices and edges to facilitate accurate and efficient end-to-end vectorized HD map learning. Furthermore, we introduce an association strategy using a Graph Neural Network to efficiently handle the complex geometry of various map elements, while maintaining scalability. Comprehensive experiments on public open dataset show that our proposed network outperforms state-of-the-art model by 1.6 mAP. We further showcase the superior scalability of our approach compared to state-of-the-art methods, achieving a 4.8 mAP improvement in long range configuration. Our code is available at https://***/juyebshin/InstaGraM. Copyright © 2023, The Authors. All rights reserved.

关键词： Graph neural networks