检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Zhou, Xuan Shi, Xiang Zhang, Lele Chen, Chen Li, Hongbo Ma, Lin Deng, Fang Chen, Jie The State Key Lab of Autonomous Intelligent Unmanned Systems School of Automation Beijing Institute of Technology Beijing100081 China The Department of Automation Tsinghua University Beijing10084 China The Beijing Geek+ Technology Co. Ltd Beijing100000 China Zhejiang Cainiao Supply Chain Management Company Ltd. Hangzhou311101 China The State Key Lab of Autonomous Intelligent Unmanned Systems Beijing Institute of Technology Beijing100081 China Shanghai Research Institute for Intelligent Autonomous Systems Tongji University Shanghai200092 China

To improve the efficiency of warehousing system and meet huge customer orders, we aim to solve the challenges of dimension disaster and dynamic properties in hyper scale multirobot task planning (MRTP) for robotic mobile fulfillment system (RMFS). Existing research indicates that hierarchical reinforcement learning (HRL) is an effective method to reduce these challenges. Based on that, we construct an efficient multi-stage HRL-based multi-robot task planner for hyper scale MRTP in RMFS, and the planning process is represented with a special temporal graph topology. Following its temporal logic, only critical events deserve attention, so system can sample efficiently when training and hold dynamic response in execution. To ensure optimality, the planner is designed with a centralized architecture, but it also brings the challenges of scaling up and generalization that require policies to maintain performance for various unlearned scales and maps. To tackle these difficulties, we first construct a hierarchical temporal attention network (HTAN) to ensure basic ability of handling inputs with unfixed lengths, and then design multi-stage curricula for hierarchical policy learning to further improve the scaling up and generalization ability while avoiding catastrophic forgetting. Additionally, we notice that policies with hierarchical structure suffer from unfair credit assignment that is similar to that in multi-agent reinforcement learning, inspired of which, we propose a hierarchical reinforcement learning algorithm with counterfactual rollout baseline to improve learning performance. Experimental results demonstrate that our planner outperform other state-of-the-art methods on various MRTP instances in both simulated and real-world RMFS. Also, our planner can successfully scale up to hyper scale MRTP instances in RMFS with up to 200 robots and 1000 retrieval racks on unlearned maps while keeping superior performance over other methods. Copyright © 2024, The Authors. A

关键词： Multipurpose robots

来源：评论

学校读者我要写书评

暂无评论

Negative Conductivity Induced Reconfigurable Gain Metasurfaces and Their Nonlinearity

引用

Physical Review Letters 2024年第11期133卷 113801页

作者： Xiaoyue Zhu Chao Qian Erping Li Hongsheng Chen ZJU-UIUC Institute Interdisciplinary Center for Quantum Information State Key Laboratory of Extreme Photonics and Instrumentation ZJU-Hangzhou Global Science and Technology Innovation Center Key Lab. of Advanced Micro/Nano Electronic Devices & Smart Systems of Zhejiang Jinhua Institute of Zhejiang University

The past decades have witnessed the rapid development of metamaterials and metasurfaces. However, loss is still a challenging problem limiting numerous practical applications, including long-range wireless communications, superscattering, and non-Hermitian physics. Recently, great effort has been made to minimize the loss, however, they are too complicated for practical implementation and still restricted by the theoretical limit. Here, we propose and experimentally realize a tunable gain metasurface induced by negative conductivity, with deep theoretical analysis from scattering theory and equivalent circuits. In the experiment, we create metasurface samples embedded with tunable negative (or positive) conductivity to achieve adjustable gain (or loss). By varying the control bias voltages, the metasurfaces can reflect incident waves with additional controllab.e gain. Interestingly, we find the gain metasurfaces inherently pose nonlinearities, which are beneficial for nonlinear optics and microwave applications, particularly for the nonlinear activation of wave-based neural networks.

关键词： Electromagnetic waves & oscillations Gain properties of lasers Metamaterials Metasurfaces Nonlinear optics

来源：评论

学校读者我要写书评

暂无评论

Isotropic3D: Image-to-3D Generation Based on a Single CLIP Embedding

arXiv

引用

arXiv 2024年

作者： Liu, Pengkun Wang, Yikai Sun, Fuchun Li, Jiafang Xiao, Hang Xue, Hongxiang Wang, Xinzhou Academy for Engineering and Technology Fudan University Shanghai China State Key Lab on Intelligent Technology and Systems Department of Computer Science and Technology Tsinghua University Beijing China Tongji University Shanghai China ShengShu Beijing China

Encouraged by the growing availab.lity of pre-trained 2D diffusion models, image-to-3D generation by leveraging Score Distillation Sampling (SDS) is making remarkable progress. Most existing methods combine novel-view lifting from 2D diffusion models which usually take the reference image as a condition while applying hard L2 image supervision at the reference view. Yet heavily adhering to the image is prone to corrupting the inductive knowledge of the 2D diffusion model leading to flat or distorted 3D generation frequently. In this work, we reexamine image-to-3D in a novel perspective and present Isotropic3D, an image-to-3D generation pipeline that takes only an image CLIP embedding as input. Isotropic3D allows the optimization to be isotropic w.r.t. the azimuth angle by solely resting on the SDS loss. The core of our framework lies in a two-stage diffusion model fine-tuning. Firstly, we fine-tune a text-to-3D diffusion model by substituting its text encoder with an image encoder, by which the model preliminarily acquires image-to-image capabilities. Secondly, we perform fine-tuning using our Explicit Multi-view Attention (EMA) which combines noisy multi-view images with the noise-free reference image as an explicit condition. CLIP embedding is sent to the diffusion model throughout the whole process while reference images are discarded once after fine-tuning. As a result, with a single image CLIP embedding, Isotropic3D is capable of generating multi-view mutually consistent images and also a 3D model with more symmetrical and neat content, well-proportioned geometry, rich colored texture, and less distortion compared with existing image-to-3D methods while still preserving the similarity to the reference image to a large extent. The project page is availab.e at https://***/. The code and models are availab.e at https://***/pkunliu/Isotropic3D. Copyright © 2024, The Authors. All rights reserved.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

A DESIGN OF BOX-METER BINDING SYSTEM FOR POWER METERING BOX 22

A DESIGN OF BOX-METER BINDING SYSTEM FOR POWER METERING BOX

引用

3rd International Conference on Control, Robotics and intelligent System, CCRIS 2022

作者： He, Xing Huang, Rui Yu, Minqi Zeng, Wenwei Zhang, Suihan State Grid Hunan Electric Power Company Limited Hunan Province Key Laboratory of Intelligent Electrical Measurement and Application Technology Changsha Hunan410000 China College of Electrical and Information Engineering Hunan University Changsha Hunan410000 China State Grid Hunan Electric Power Company Limited China and Hunan Prov. Key Lab. of Intelligent Electrical Measurement and Application Technology China

ISBN: (纸本)9781450396851

At present, most of the topological identification and binding methods of box tables in power metering boxes are supported by the later platform algorithm. When installing or replacing electricity meters is installed, it is easy to lead to the wrong identification of box-meter binding. In this paper, the MCU and the multi-channel RFID identification board to obtain the asset information in the meter lab.l, record the asset information data analysis and forward, and realize the accurate search, positioning, identification, recording, binding and other functions in the metering box. After the realization of the table and box relationship binding, the system establishes the relationship table internally, uploads the box and table relationship table to the energy controller through HPLC (High Performance Liquid Chromatography), and establishes the box and table ledger data of each metering box in the platform area. Through this method, the binding relationship between the individual meter and the corresponding position in the box is realized, which solves the problems of the meter management chaos and the box meter position relationship in the operation and maintenance process, which is of great significance to the humanized and efficient development of smart meters. © 2022 ACM.

关键词： Radio frequency identification (RFID)

来源：评论

学校读者我要写书评

暂无评论

Small Scale Data-Free Knowledge Distillation

arXiv

引用

arXiv 2024年

作者： Liu, He Wang, Yikai Liu, Huaping Sun, Fuchun Yao, Anbang China Telecom Cloud Technology Co. Ltd. China BNRist Center State Key Lab on Intelligent Technology and Systems Department of Computer Science and Technology Tsinghua University China Intel Labs China China

Data-free knowledge distillation is able to utilize the knowledge learned by a large teacher network to augment the training of a smaller student network without accessing the original training data, avoiding privacy, security, and proprietary risks in real applications. In this line of research, existing methods typically follow an inversion-and-distillation paradigm in which a generative adversarial network on-the-fly trained with the guidance of the pre-trained teacher network is used to synthesize a large-scale sample set for knowledge distillation. In this paper, we reexamine this common data-free knowledge distillation paradigm, showing that there is considerable room to improve the overall training efficiency through a lens of "small-scale inverted data for knowledge distillation". In light of three empirical observations indicating the importance of how to balance class distributions in terms of synthetic sample diversity and difficulty during both data inversion and distillation processes, we propose Small Scale Data-free Knowledge Distillation (SSD-KD). In formulation, SSD-KD introduces a modulating function to balance synthetic samples and a priority sampling function to select proper samples, facilitated by a dynamic replay buffer and a reinforcement learning strategy. As a result, SSD-KD can perform distillation training conditioned on an extremely small scale of synthetic samples (e.g., 10× less than the original training data scale), making the overall training efficiency one or two orders of magnitude faster than many mainstream methods while retaining superior or competitive model performance, as demonstrated on popular image classification and semantic segmentation benchmarks. The code is availab.e at https://***/OSVAI/SSD-KD. Copyright © 2024, The Authors. All rights reserved.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Distributed Online Convex Optimization with Time-Varying Constraints: Tighter Cumulative Constraint Violation Bounds under Slater's Condition

引用

IEEE Transactions on Automatic Control 2025年

作者： Yi, Xinlei Li, Xiuxian Yang, Tao Xie, Lihua Hong, Yiguang Chai, Tianyou Johansson, Karl H. National Key Laboratory of Autonomous Intelligent Unmanned Systems Shanghai Institute of Intelligent Science and Technology China Ministry of Education Frontiers Science Center for Intelligent Autonomous Systems China Massachusetts Institute of Technology Lab for Information & Decision Systems CambridgeMA02139 United States Northeastern University State Key Laboratory of Synthetical Automation for Process Industries Shenyang110819 China Nanyang Technological University School of Electrical and Electronic Engineering 50 Nanyang Avenue 639798 Singapore Kth Royal Institute of Technology Division of Decision and Control Systems School of Electrical Engineering and Computer Science Sweden Digital Futures Stockholm10044 Sweden

This paper considers distributed online convex optimization with time-varying constraints. In this setting, a network of agents makes decisions at each round, and then only a portion of the loss function and a coordinate block of the constraint function are privately revealed to each agent. The loss and constraint functions are convex and can vary arbitrarily across rounds. The agents collab.rate to minimize static network regret and network cumulative constraint violation. A novel distributed online algorithm with a vanishing stepsize is proposed and it achieves an O(Tmax {c,1-c}) static network regret bound and an O(T1-c/2) network cumulative constraint violation bound, where T is the number of rounds and c∈ (0,1) is a user-defined trade-off parameter. When Slater's condition holds (i.e, there is a point that strictly satisfies the inequality constraints), the network cumulative constraint violation bound is reduced to O(T1-c). Moreover, if the loss functions are strongly convex, then static network regret bound is reduced to O(log (T)), and the network cumulative constraint violation bound is reduced to O(√log (T)T) and O(log (T)) without and with Slater's condition, respectively. To the best of our knowledge, this paper is the first to achieve tighter (network) cumulative constraint violation bounds for (distributed) online convex optimization with time-varying constraints under Slater's condition. Finally, the theoretical results are verified through numerical simulations. © 1963-2012 IEEE.

关键词： Convex optimization

来源：评论

学校读者我要写书评

暂无评论

COLD: A Benchmark for Chinese Offensive Language Detection Disclaimer: The paper contains content that may be profane, vulgar, or offensive

COLD: A Benchmark for Chinese Offensive Language Detection D...

引用

2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022

作者： Deng, Jiawen Zhou, Jingyan Sun, Hao Zheng, Chujie Mi, Fei Meng, Helen Huang, Minlie The CoAI group DCST Institute for Artificial Intelligence State Key Lab of Intelligent Technology and Systems China Beijing National Research Center for Information Science and Technology Tsinghua University Beijing100084 China Dept. of Systems Engineering & Engineering Management The Chinese University of Hong Kong Hong Kong Huawei Noah's Ark Lab

Offensive language detection is increasingly crucial for maintaining a civilized social media platform and deploying pre-trained language models. However, this task in Chinese is still under exploration due to the scarcity of reliable datasets. To this end, we propose a benchmark - COLD for Chinese offensive language analysis, including a Chinese Offensive Language Dataset - COLDATASET and a baseline detector - COLDETECTOR which is trained on the dataset. We show that the COLD benchmark contributes to Chinese offensive language detection which is challenging for existing resources. We then deploy the COLDETECTOR and conduct detailed analyses on popular Chinese pre-trained language models. We first analyze the offensiveness of existing generative models and show that these models inevitably expose varying degrees of offensive issues. Furthermore, we investigate the factors that influence the offensive generations, and we find that anti-bias contents and keywords referring to certain groups or revealing negative attitudes trigger offensive outputs easier. © 2022 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

AMOR: A Recipe for Building Adaptable Modular Knowledge Agents Through Process Feedback

arXiv

引用

arXiv 2024年

作者： Guan, Jian Wu, Wei Wen, Zujie Xu, Peng Wang, Hongning Huang, Minlie The CoAI Group DCST 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 Ant Group China

The notable success of large language models (LLMs) has sparked an upsurge in building language agents to complete various complex tasks. We present AMOR, an agent framework based on open-source LLMs, which reasons with external knowledge bases and adapts to specific domains through human supervision to the reasoning process. AMOR builds reasoning logic over a finite state machine (FSM) that solves problems through autonomous executions and transitions over disentangled modules. This allows humans to provide direct feedback to the individual modules, and thus naturally forms process supervision. Based on this reasoning and feedback framework, we develop AMOR through two-stage fine-tuning: warm-up and adaptation. The former fine-tunes the LLM with examples automatically constructed from various public datasets, enabling AMOR to generalize across different knowledge environments, while the latter tailors AMOR to specific domains using process feedback. Extensive experiments across multiple domains demonstrate the advantage of AMOR to strong baselines, thanks to its FSM-based reasoning and process feedback mechanism. The code and data are publicly availab.e at https://***/JianGuanTHU/AMOR. Copyright © 2024, The Authors. All rights reserved.

关键词： Open source software

来源：评论

学校读者我要写书评

暂无评论

Automatic Comment Generation for Chinese Student Narrative Essays

Automatic Comment Generation for Chinese Student Narrative E...

引用

2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022

作者： Zhang, Zhexin Guan, Jian Xu, Guowei Tian, Yixiang Huang, Minlie The CoAI Group DCST China Institute for Artificial Intelligence United States State Key Lab of Intelligent Technology and Systems China Beijing National Research Center for Information Science and Technology China Tsinghua University Beijing100084 China TAL Education Group China

ISBN: (纸本)9781959429418

Automatic essay evaluation can help reduce teachers' workload and enable students to refine their works rapidly. Previous studies focus mainly on giving discrete scores for either the holistic quality or several distinct traits. However, real-world teachers usually provide detailed comments in natural language, which are more informative than single scores. In this paper, we present the comment generation task, which aims to generate comments for specified segments from given student narrative essays. To tackle this task, we propose a planning-based generation model, which first plans a sequence of keywords, and then expands these keywords into a complete comment. To improve the correctness and informativeness of generated comments, we adopt two following techniques: (1) training an error correction module to filter out incorrect keywords, and (2) recognizing fine-grained structured features from source essays to enrich the key word. To support the evaluation of the task, we collect a human-written Chinese dataset, which contains 22,399 essay-comment pairs. Extensive experiments show that our model outperforms strong baselines significantly. Moreover, we exert explicit control on our model to generate comments to describe the strengths or weaknesses of inputs with a 91% success rate. We deploy the model at http://***/static/essayComment/. A demo video is availab.e at https://***/IuFVk8dUxbI. Our code and data are availab.e at https://***/thu-coai/EssayCommentGen. © 2022 Association for Computational Linguistics.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Deep multimodal fusion by channel exchanging 34

Deep multimodal fusion by channel exchanging

引用

34th Conference on Neural Information Processing systems, NeurIPS 2020

作者： Wang, Yikai Huang, Wenbing Sun, Fuchun Xu, Tingyang Rong, Yu Huang, Junzhou State Key Lab on Intelligent Technology and Systems Department of Computer Science and Technology Tsinghua University China Tencent AI Lab.

Deep multimodal fusion by using multiple sources of data for classification or regression has exhibited a clear advantage over the unimodal counterpart on various applications. Yet, current methods including aggregation-based and alignment-based fusion are still inadequate in balancing the trade-off between inter-modal fusion and intra-modal processing, incurring a bottleneck of performance improvement. To this end, this paper proposes Channel-Exchanging-Network (CEN), a parameter-free multimodal fusion framework that dynamically exchanges channels between sub-networks of different modalities. Specifically, the channel exchanging process is self-guided by individual channel importance that is measured by the magnitude of Batch-Normalization (BN) scaling factor during training. The validity of such exchanging process is also guaranteed by sharing convolutional filters yet keeping separate BN layers across modalities, which, as an add-on benefit, allows our multimodal architecture to be almost as compact as a unimodal network. Extensive experiments on semantic segmentation via RGB-D data and image translation through multi-domain input verify the effectiveness of our CEN compared to current state-of-the-art methods. Detailed ablation studies have also been carried out, which provably affirm the advantage of each component we propose. Our code is availab.e at https://***/yikaiw/CEN. © 2020 Neural information processing systems foundation. All rights reserved.

关键词： Semantics

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：