检索结果-内蒙古大学图书馆

SHARING KNOWLEDGE IN MULTI-TASK DEEP REINFORCEMENT LEARNING

学校读者我要写书评

暂无评论

arXiv 2024年

作者： D'Eramo, Carlo Tateo, Davide Bonarini, Andrea Restelli, Marcello Peters, Jan Department of Computer Science TU Darmstadt IAS Hochschulstraße 10 Darmstadt64289 Germany Politecnico di Milano DEIB Piazza Leonardo da Vinci 32 Milano20133 Italy TU Darmstadt IAS Hochschulstraße 10 Darmstadt64289 Germany Max Planck Institute for Intelligent Systems Max-Planck-Ring 4 Tübingen72076 Germany

We study the benefit of sharing representations among tasks to enable the effective use of deep neural networks in Multi-Task Reinforcement Learning. We leverage the assumption that learning from different tasks, sharing common properties, is helpful to generalize the knowledge of them resulting in a more effective feature extraction compared to learning a single task. Intuitively, the resulting set of features offers performance benefits when used by Reinforcement Learning algorithms. We prove this by providing theoretical guarantees that highlight the conditions for which is convenient to share representations among tasks, extending the well-known finite-time bounds of Approximate Value-Iteration to the multi-task setting. In addition, we complement our analysis by proposing multi-task extensions of three Reinforcement Learning algorithms that we empirically evaluate on widely used Reinforcement Learning benchmarks showing significant improvements over the single-task counterparts in terms of sample efficiency and performance. © 2024, CC BY.

关键词： Reinforcement learning

On Embeddings and Inverse Embeddings of Input Design for Regularized System Identification

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Mu, Biqiang Chen, Tianshi Kong, He Jiang, Bo Wang, Lei Wu, Junfeng Key Laboratory of Systems and Control Institute of Systems Science Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing100190 China School of Data Science Shenzhen Research Institute of Big Data The Chinese University of Hong Kong Shenzhen518172 China Shenzhen Key Laboratory of Biomimetic Robotics and Intelligent Systems Department of Mechanical and Energy Engineering Southern University of Science and Technology Shenzhen518055 China Guangdong Provincial Key Laboratory of Human-Augmentation and Rehabilitation Robotics in Universities Southern University of Science and Technology Shenzhen518055 China Key Laboratory for NSLSCS of Jiangsu Province School of Mathematical Sciences Nanjing Normal University Nanjing China College of Control Science and Engineering Zhejiang University Hangzhou China

Input design is an important problem for system identification and has been well studied for the classical system identification, i.e., the maximum likelihood/prediction error method. For the emerging regularized system identification, the study on input design has just started, and it is often formulated as a non-convex optimization problem that minimizes a scalar measure of the Bayesian mean squared error matrix subject to certain constraints, and the state-of-art method is the so-called quadratic mapping and inverse embedding (QMIE) method, where a time domain inverse embedding (TDIE) is proposed to find the inverse of the quadratic mapping. In this paper, we report some new results on the embeddings/inverse embeddings of the QMIE method. Firstly, we present a general result on the frequency domain inverse embedding (FDIE) that is to find the inverse of the quadratic mapping described by the discrete-time Fourier transform. Then we show the relation between the TDIE and the FDIE from a graph signal processing perspective. Finally, motivated by this perspective, we further propose a graph induced embedding and its inverse, which include the previously introduced embeddings as special cases. This deepens the understanding of input design from a new viewpoint beyond the real domain and the frequency domain viewpoints. Copyright © 2022, The Authors. All rights reserved.

关键词： Embeddings

D-Grasp: Physically Plausible Dynamic Grasp Synthesis for Hand-Object Interactions

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Christen, Sammy Kocabas, Muhammed Aksan, Emre Hwangbo, Jemin Song, Jie Hilliges, Otmar Department of Computer Science ETH Zurich Switzerland Max Planck Institute for Intelligent Systems Tübingen Germany Department of Mechanical Engineering KAIST Korea Republic of

We introduce the dynamic grasp synthesis task: given an object with a known 6D pose and a grasp reference, our goal is to generate motions that move the object to a target 6D pose. This is challenging, because it requires reasoning about the complex articulation of the human hand and the intricate physical interaction with the object. We propose a novel method that frames this problem in the reinforcement learning framework and leverages a physics simulation, both to learn and to evaluate such dynamic interactions. A hierarchical approach decomposes the task into low-level grasping and high-level motion synthesis. It can be used to generate novel hand sequences that approach, grasp, and move an object to a desired location, while retaining human-likeness. We show that our approach leads to stable grasps and generates a wide range of motions. Furthermore, even imperfect labels can be corrected by our method to generate dynamic interaction sequences. Video and code are available at: https://***/d-grasp/. Copyright © 2021, The Authors. All rights reserved.

关键词： Dynamics

Multi-scale Feature Fusion with Point Pyramid for 3D Object Detection

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Lu, Weihao Zhao, Dezong Premebida, Cristiano Zhang, Li Zhao, Wenjing Tian, Daxin The School of Engineering University of Glasgow University Avenue GlasgowG12 8QQ United Kingdom The Institute of Systems and Robotics University of Coimbra Coimbra3030-290 Portugal The Department of Computer Science Royal Holloway University of London SurreyTW20 0EX United Kingdom The Department of Civil Environmental Engineering Hong Kong Polytechnic University Hong Kong The School of Transportation Science and Engineering Beihang University Haidian District Beijing100191 China

Effective point cloud processing is crucial to LiDAR-based autonomous driving systems. The capability to understand features at multiple scales is required for object detection of intelligent vehicles, where road users may appear in different sizes. Recent methods focus on the design of the feature aggregation operators, which collect features at different scales from the encoder backbone and assign them to the points of interest. While efforts are made into the aggregation modules, the importance of how to fuse these multi-scale features has been overlooked. This leads to insufficient feature communication across scales. To address this issue, this paper proposes the Point Pyramid RCNN (POP-RCNN), a feature pyramid-based framework for 3D object detection on point clouds. POP-RCNN consists of a Point Pyramid Feature Enhancement (PPFE) module to establish connections across spatial scales and semantic depths for information exchange. The PPFE module effectively fuses multi-scale features for rich information without the increased complexity in feature aggregation. To remedy the impact of inconsistent point densities, a point density confidence module is deployed. This design integration enables the use of a lightweight feature aggregator, and the emphasis on both shallow and deep semantics, realising a detection framework for 3D object detection. With great adaptability, the proposed method can be applied to a variety of existing frameworks to increase feature richness, especially for long-distance detection. By adopting the PPFE in the voxel-based and point-voxel-based baselines, experimental results on KITTI and Waymo Open Dataset show that the proposed method achieves remarkable performance even with limited computational headroom. © 2024, CC BY-NC-ND.

关键词： Object detection

Sim2Real Object-Centric Keypoint Detection and Description

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zhong, Chengliang Yang, Chao Qi, Jinshan Sun, Fuchun Liu, Huaping Mu, Xiaodong Huang, Wenbing Xi'an Research Institute of High-Tech Xi'An710025 China State Key Lab on Intelligent Technology and Systems Department of Computer Science and Technology Tsinghua University Beijing100084 China Institute for AI Industry Research Tsinghua University China Shandong University of Science and Technology Qingdao266590 China

Keypoint detection and description play a central role in computer vision. Most existing methods are in the form of scene-level prediction, without returning the object classes of different keypoints. In this paper, we propose the object-centric formulation, which, beyond the conventional setting, requires further identifying which object each interest point belongs to. With such fine-grained information, our framework enables more downstream potentials, such as object-level matching and pose estimation in a clustered environment. To get around the difficulty of label collection in the real world, we develop a sim2real contrastive learning mechanism that can generalize the model trained in simulation to real-world applications. The novelties of our training method are three-fold: (i) we integrate the uncertainty into the learning framework to improve feature description of hard cases, e.g., less-textured or symmetric patches;(ii) we decouple the object descriptor into two output branches-intra-object salience and inter-object distinctness, resulting in a better pixel-wise description;(iii) we enforce cross-view semantic consistency for enhanced robustness in representation learning. Comprehensive experiments on image matching and 6D pose estimation verify the encouraging generalization ability of our method from simulation to reality. Particularly for 6D pose estimation, our method significantly outperforms typical unsupervised/sim2real methods, achieving a closer gap with the fully supervised counterpart. Additional results and videos can be found at https://***/rock/. Copyright © 2022, The Authors. All rights reserved.

关键词： Object detection

Enhanced wave localization in multifractal scattering media

学校读者我要写书评

暂无评论

Physical Review B 2023年第5期107卷 054201-054201页

作者： Y. Chen F. Sgrignuoli Y. Zhu T. Shubitidze L. Dal Negro Department of Electrical and Computer Engineering & Photonics Center Boston University 8 Saint Mary's Street Boston Massachusetts 02215 USA Institute of Applied Sciences and Intelligent Systems National Research Council Via Pietro Castellino Naples 80131 Italy Division of Materials Science & Engineering Boston University 15 Saint Mary's St. Brookline Massachusetts 02446 USA Department of Physics Boston University 590 Commonwealth Avenue Boston Massachusetts 02215 USA

In this paper, we study the structural, scattering, and wave localization properties of multifractal arrays of electric point dipoles generated from multiplicative random fields with different spectra of fractal dimensions. Specifically, using the rigorous Green's matrix method, we investigate the scattering resonances and wave localization behavior of systems with N=104 dipoles and demonstrate an enhanced localization behavior in highly inhomogeneous multifractal structures compared to homogeneous fractals, or monofractals. We show distinctive spectral properties, such as the absence of level repulsion in the strong multiple scattering regime and power-law statistics of level spacings, which indicate a clear localization transition enhanced in nonhomogeneous multifractals. Our findings unveil the importance of multifractal structural correlations in the multiple scattering regime of electric dipole arrays and provide an efficient model for the design of multiscale nanophotonic systems with enhanced light-matter coupling and localization phenomena beyond what is possible with traditional fractal systems.

关键词： Anderson localization Classical optics Metamaterials Scattering theory

Regression Models for Predicting Gamified Educational Outcomes

学校读者我要写书评

暂无评论

Regression Models for Predicting Gamified Educational Outcom...

Telecommunications Conference (ITC-Egypt), International

作者： Abdelhameed Ibrahim Christos Gatzoulis Marwa M. Eid Faris H. Rizk Laith Abualigah El-Sayed M. El-kenawy Faculty of Engineering Design and Information & Communications Technology (EDICT) School of ICT Bahrain Polytechnic Isa Town Bahrain Faculty of Artificial Intelligence Delta University for Science and Technology Mansoura Egypt Computer Science and Intelligent Systems Research Center Blacksburg Virginia USA Hourani Center for Applied Scientific Research Al-Ahliyya Amman University Amman Jordan Computer Science Department Al al-Bayt University Mafraq Jordan MEU Research Unit Middle East University Amman Jordan Applied Science Research Center Applied Science Private University Amman Jordan Department of Communications and Electronics Delta Higher Institute of Engineering and Technology Mansoura Egypt

ISBN: (数字)9798350351408

ISBN: (纸本)9798350351415

This paper focuses on improving learning and involvement in studies by employing game elements and considering online learning platforms to deliver statistics curriculum. According to the problem statement, the research examines gamification methods to transform studying from a routine to having enjoyment while improving academic results. This area of investigation is being extended to predict academic outcomes by applying different machine learning models to student performance data received from gamified. The strategy entails using a pedagogy based on collecting data from students who participated in a gamified educational platform; the data comprises indicators like pre-and post-platform exam grades, grades on quizzes, and indicators of platform usage. Multiple regression algorithms, including Random Forest and Gradient Boosting, are applied to student performance. In the experiments, the output results indicate that the model of Random Forest Regressor outperforms other models with the least Mean Squared Error (MSE) of 0.052. This model performs the best in terms of prediction accuracy with competence level in predicting students' results in gamified environments. The findings can support decision-making in designing educational activities. The Random Forest Regressor model can be identified as the best predictive model and may help train gamified learning platforms, which can encourage student engagement that correlates to performance.

关键词： Machine learning algorithms Transforms Predictive models Nearest neighbor methods Prediction algorithms Boosting Data models

How Can Businesses, Low-Tech Businesses in Particular, Benefit from the Metaverse? A Delphi Perspective from Academics

学校读者我要写书评

暂无评论

How Can Businesses, Low-Tech Businesses in Particular, Benef...

International Conference on Marketing and Technologies, ICMarkTech 2023

作者： Lopes, Teresa Sofia Amorim Carreira, Rui Carlos, Vera Silva Pereira, Rui Humberto Alves, Carlos Pratas, Joaquim dos Santos, José Paulo Marques Department of Business Administration University of Maia Maia Portugal NIPE—Centre for Research in Economics and Management School of Economics and Management University of Minho Braga Portugal Polytechnic Institute of Maia Maia Portugal LIACC—Artificial Intelligence and Computer Science Laboratory Faculty of Engineering University of Porto Porto Portugal NECE—Research Center for Business Sciences Dep. of Management and Economics University of Beira Interior Covilhã Portugal CEOS.PP—Centre for Organizational and Social Studies Polytechnic of Porto Porto Portugal Algoritmi Research Centre University of Minho Guimarães Portugal LASI—Intelligent Systems Associate Laboratory Guimarães Portugal Porto Accounting and Business School Polytechnic Institute of Porto São Mamede de Infesta Portugal

ISBN: (纸本)9789819715510

Two years ago, Meta was born, accompanied by the announcement of the Metaverse. The Metaverse would be a collection of 3D immersive ecosystems for people’s socialization. This innovative idea should allow one to live in the Metaverse as in real life without physical barriers. Attending shows, concerts, virtual travel, and caring for one’s health or education should be possible. In sum, a new marketplace is born. Amid such newness, the question of how businesses may benefit from the Metaverse arises. Albeit tech businesses must be comfortable and adapted to the potentialities of the new marketplace, the entrance of the low-techs may be more problematic. The study brings together a panel of academic experts to reflect on these questions. They agree that there is a myriad of potentialities in the new marketplace, not only for B2C but also for B2B. Low-techs may be less adapted to benefit from the Metaverse. A following-the-wave strategy must be avoided. On the contrary, investing in understanding the new paradigm and careful planning is essential. The Metaverse has no physical barriers, which tend to limit low-techs’ actions. Their nonexistence creates new opportunities and must be where customers are, after all. The Metaverse, however, still needs development to gain critical mass. Hardware easy to use, its cost, speedy and cheap Internet, and mainly, trustable and recognized currencies are still required. Nonetheless, the Metaverse may be a new world full of opportunities. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Commerce

Neural-iLQR: A Learning-Aided Shooting Method for Trajectory Optimization

学校读者我要写书评

暂无评论

Neural-iLQR: A Learning-Aided Shooting Method for Trajectory...

IEEE International Conference on robotics and Biomimetics

作者： Zilong Cheng Yulin Li Kai Chen Jianghua Duan Jun Ma Tong Heng Lee Department of Electrical and Computer Engineering National University of Singapore Singapore Division of Emerging Interdisciplinary Areas The Hong Kong University of Science and Technology Hong Kong SAR China Robotics and Autonomous Systems Thrust The Hong Kong University of Science and Technology (Guangzhou) Guangzhou China Department of Mechanical and Aerospace Engineering The Hong Kong University of Science and Technology Hong Kong SAR China HKUST Shenzhen-Hong Kong Collaborative Innovation Research Institute Shenzhen China

Iterative linear quadratic regulator (iLQR) has gained wide popularity in addressing trajectory optimization problems with nonlinear system models. However, as a model-based shooting method, it relies heavily on an accurate system model to update the optimal control actions and the trajectory determined with forward integration, thus becoming vulnerable to inevitable model inaccuracies. Recently, substantial research efforts in learning-based methods for optimal control problems have been progressing significantly in addressing unknown system models, particularly when the system has complex interactions with the environment. Yet an incremental machine learning method or a deep neural network is normally required to fit a substantial scale of sampling motion data. Existing works mainly focus on capturing the whole picture of the motion pattern from trial data (which is essentially noisy and hard to identify with any simple predefined structures) for complex robot systems. Instead, we present Neural-iLQR, a learning-aided shooting method over the unconstrained control space, in which a neural network with a simple structure is used to drive the optimization process towards optimal direction. Through comprehensive evaluations on two illustrative control tasks using three different kinds of lightweight network structures, the proposed method is shown to outperform the conventional iLQR significantly with fast and continuous cost convergence in the presence of inaccuracies in system models, which demonstrates the generalizability and robustness of the proposed learning-aided method.

关键词：