检索结果-内蒙古大学图书馆

9th International Conference on Learning Representations, ICLR 2021

作者： Patel, Kanil Beluch, William Yang, Bin Pfeiffer, Michael Zhang, Dan Bosch Center for Artificial Intelligence Renningen Germany Institute of Signal Processing and System Theory University of Stuttgart Stuttgart Germany

Post-hoc multi-class calibration is a common approach for providing high-quality confidence estimates of deep neural network predictions. Recent work has shown that widely used scaling methods underestimate their calibration error, while alternative Histogram Binning (HB) methods often fail to preserve classification accuracy. When classes have small prior probabilities, HB also faces the issue of severe sample-inefficiency after the conversion into K one-vs-rest class-wise calibration problems. The goal of this paper is to resolve the identified issues of HB in order to provide calibrated confidence estimates using only a small holdout calibration dataset for bin optimization while preserving multi-class ranking accuracy. From an information-theoretic perspective, we derive the I-Max concept for binning, which maximizes the mutual information between labels and quantized logits. This concept mitigates potential loss in ranking performance due to lossy quantization, and by disentangling the optimization of bin edges and representatives allows simultaneous improvement of ranking and calibration performance. To improve the sample efficiency and estimates from a small calibration set, we propose a shared class-wise (sCW) calibration strategy, sharing one calibrator among similar classes (e.g., with similar class priors) so that the training sets of their class-wise calibration problems can be merged to train the single calibrator. The combination of sCW and I-Max binning outperforms the state of the art calibration methods on various evaluation metrics across different benchmark datasets and models, using a small calibration set (e.g., 1k samples for ImageNet). © 2021 ICLR 2021 - 9th International Conference on Learning Representations. All rights reserved.

关键词： Calibration

来源：评论

学校读者我要写书评

暂无评论

Data-Driven Predictive Disturbance Observer for Quasi Continuum Manipulators

Data-Driven Predictive Disturbance Observer for Quasi Contin...

引用

IEEE Conference on Decision and Control

作者： Daniel Müller Justinus Feilhauer Jennifer Wickert Julian Berberich Frank Allgöwer Oliver Sawodny Institute for System Dynamics University of Stuttgart Stuttgart Institute for Systems Theory and Automatic Control University of Stuttgart Stuttgart Germany

ISBN: (数字)9781665467612

ISBN: (纸本)9781665467629

Deriving accurate and time efficient models for soft robots has proven to be a difficult task. This is mainly due to the behaviour of the soft materials and the interaction with other parts of the robot, resulting in complicated models with many states that require complex sensor *** robots such as the Bionic Soft Arm (BSA) are subject to uncertainties. The resulting mismatch between plant and model can be viewed as a disturbance. Disturbance observers significantly improve the control performance of the model-based controllers. However, they assume an underlying model of the disturbance that is usually unknown. To address this problem, we propose a novel Data-Driven Predictive Disturbance Observer (DPDO), which predicts the future disturbance based on past measurements that are updated online and, thereby, enables the model-based controller to achieve zero control error. The approach is first tested in simulation, examining the influence of noise, time delays and typical model uncertainties for the BSA. Afterwards, the algorithm is applied to the real system, demonstrating its practicability.

关键词： Uncertainty Trajectory tracking Biological system modeling Soft robotics Predictive models Prediction algorithms Manipulators

来源：评论

学校读者我要写书评

暂无评论

CMGAN: Conformer-Based Metric-GAN for Monaural Speech Enhancement

arXiv

引用

arXiv 2022年

作者： Abdulatif, Sherif Cao, Ruizhe Yang, Bin The Institute of Signal Processing and System Theory University of Stuttgart Stuttgart70569 Germany

In this work, we further develop the conformer-based metric generative adversarial network (CMGAN) model1 for speech enhancement (SE) in the time-frequency (TF) domain. This paper builds on our previous work but takes a more in-depth look by conducting extensive ablation studies on model inputs and architectural design choices. We rigorously tested the generalization ability of the model to unseen noise types and distortions. We have fortified our claims through DNSMOS measurements and listening tests. Rather than focusing exclusively on the speech denoising task, we extend this work to address the dereverberation and super-resolution tasks. This necessitated exploring various architectural changes, specifically metric discriminator scores and masking techniques. It is essential to highlight that this is among the earliest works that attempted complex TF-domain super-resolution. Our findings show that CMGAN outperforms existing state-of-the-art methods in the three major speech enhancement tasks: denoising, dereverberation, and super-resolution. For example, in the denoising task using the Voice Bank+DEMAND dataset, CMGAN notably exceeded the performance of prior models, attaining a PESQ score of 3.41 and an SSNR of 11.10 dB. Audio samples and CMGAN implementations are available online2 Copyright © 2022, The Authors. All rights reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Prompt Tuning for Parameter-efficient Medical Image Segmentation

arXiv

引用

arXiv 2022年

作者： Fischer, Marc Bartler, Alexander Yang, Bin Institute of Signal Processing and System Theory University of Stuttgart Stuttgart70550 Germany

Neural networks pre-trained on a self-supervision scheme have become the standard when operating in data rich environments with scarce annotations. As such, fine-tuning a model to a downstream task in a parameter-efficient but effective way, e.g. for a new set of classes in the case of semantic segmentation, is of increasing importance. In this work, we propose and investigate several contributions to achieve a parameter-efficient but effective adaptation for semantic segmentation on two medical imaging datasets. Relying on the recently popularized prompt tuning approach, we provide a prompt-able UNet (PUNet) architecture, that is frozen after pre-training, but adaptable throughout the network by class-dependent learnable prompt tokens. We pre-train this architecture with a dedicated dense self-supervision scheme based on assignments to online generated prototypes (contrastive prototype assignment, CPA) of a student teacher combination alongside a concurrent segmentation loss on a subset of classes. We demonstrate that the resulting neural network model is able to attenuate the gap between fully fine-tuned and parameter-efficiently adapted models on CT imaging datasets. As such, the difference between fully fine-tuned and prompt-tuned variants amounts to only 3.83 pp for the TCIA/BTCV dataset and 2.67 pp for the CT-ORG dataset in the mean Dice Similarity Coefficient (DSC, in %) while only prompt tokens, corresponding to 0.85% of the pre-trained backbone model with 6.8M frozen parameters, are adjusted. The code for this work is available on https://***/marcdcfischer/PUNet. © 2022, CC BY-NC-ND.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Mechanical Properties Inside Origami-Inspired Structures: An Overview

引用

Applied Mechanics Reviews 2025年第1期77卷 011001页

作者： Yan, Peng Huang, Hailin Meloni, Marco Li, Bing Cai, Jianguo Department of Mechanical Engineering and Automation Harbin Institute of Technology Shenzhen518052 China Harbin Institute of Technology Shenzhen518052 China Key Laboratory of C & PC Structures of Ministry of Education National Prestress Engineering Research Center Southeast University Nanjing211189 China State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin150001 China Guangdong Provincial Key Laboratory of Intelligent Morphing Mechanisms and Adaptive Robotics Harbin Institute of Technology Shenzhen518052 China Key University Laboratory of Mechanism Machine Theory and Intelligent Unmanned Systems of Guangdong Habin Institute of Technology Shenzhen518052 China

In recent decades, origami has transitioned from a traditional art form into a systematic field of scientific inquiry, characterized by attributes such as high foldability, lightweight frameworks, diverse deformation modes, and limited degrees-of-freedom. Despite the abundant literature on smart materials, actuation methods, design principles, and manufacturing techniques, comprehensive reviews focusing on the mechanical properties of origami-inspired structures are relatively rare and unsystematic. This review aims to fill this void by analyzing and summarizing the significant studies conducted on the mechanical properties of origami-inspired structures from 2013 to 2023. We begin with an overview that includes essential definitions of origami, classical origami patterns, and their associated tessellated or stacked structures. Following this, we delve into the principal dynamic modeling method for origami and conduct an in-depth analysis of the key mechanical properties of origami-inspired structures. These properties include tunable stiffness, bistability and multistability, metamechanical properties demonstrated by origami-based metamaterials, and bio-inspired mechanical properties. Finally, we conclude with a comprehensive summary that discusses the current challenges and future directions in the field of origami-inspired structures. Our review provides a thorough synthesis of both the mechanical properties and practical applications of origami-inspired structures, aiming to serve as a reference and stimulate further research. © 2025 American Society of Mechanical Engineers (ASME). All rights reserved.

关键词： Degrees of freedom (mechanics)

来源：评论

学校读者我要写书评

暂无评论

Derivative-Free Data-Driven Control of Continuous-Time Linear Time-Invariant systems

arXiv

引用

arXiv 2024年

作者： Bosso, Alessandro Borghesi, Marco Iannelli, Andrea Notarstefano, Giuseppe Teel, Andrew R. Department of Electrical Electronic and Information Engineering University of Bologna Italy Institute for System Theory and Automatic Control University of Stuttgart Germany Department of Electrical and Computer Engineering University of California Santa BarbaraCA United States

This paper develops a data-driven stabilization method for continuous-time linear time-invariant systems with theoretical guarantees and no need for signal derivatives. The framework, based on linear matrix inequalities (LMIs), is illustrated in the state-feedback and single-input single-output output-feedback scenarios. Similar to discrete-time approaches, we rely solely on input and state/output measurements. To avoid differentiation, we employ low-pass filters of the available signals that, rather than approximating the derivatives, reconstruct a non-minimal realization of the plant. With access to the filter states and their derivatives, we can solve LMIs derived from sample batches of the available signals to compute a dynamic controller that stabilizes the plant. The effectiveness of the framework is showcased through numerical examples. © 2024, CC BY.

关键词： Continuous time systems

来源：评论

学校读者我要写书评

暂无评论

Tracking Control in Presence of Obstacles and Uncertainties for Bioinspired Spherical Underwater Robots

引用

Journal of Bionic engineering 2023年第1期20卷 323-337页

作者： Chunying Li Shuxiang Guo Jian Guo Graduate School of Engineering Kagawa UniversityTakamatsuKagawa761-0396Japan Key Laboratory of Convergence Medical Engineering System and Healthcare Technology the Ministry of Industry and Information TechnologySchool of Life Science and TechnologyBeijing Institute of TechnologyBeijing100081China Tianjin Key Laboratory for Control Theory and Applications in Complicated Systems and Intelligent Robot Laboratory Tianjin University of TechnologyBinshui Xidao Extension 391Tianjin300384China

During marine missions,AUVs are susceptible to external disturbances,such as obstacles,ocean currents,etc.,which can easily cause mission failure or *** this paper,considering the strong nonlinearities,external disturbances and obstacles,the kinematic and dynamic model of bioinspired Spherical Underwater Robot(SUR)was ***,the waypoints-based trajectory tracking with obstacles and uncertainties was proposed for SUR to guarantee its safety and ***,the Lyapunov theory was adopted to verify the stability and the Slide Mode Control(SMC)method is used to verify the robustness of the control *** addition,a series of simulations were conducted to evaluate the effectiveness of proposed control *** tests,including path-following,static and moving obstacle avoidance were performed which verified the feasibility,robustness and effectiveness of the designed control ***,a series of experiments in real environment were performed to verify the performance of the control *** simulation and experimental results of the study supplied clues to the improvement of the path following capability and multi-obstacle avoidance of AUVs.

关键词： Path following Obstacle avoidance Bioinspired.Spherical underwater robot(SUR) Lyapunov theory

来源：评论

学校读者我要写书评

暂无评论

CMGAN: Conformer-Based Metric-GAN for Monaural Speech Enhancement

TechRxiv

引用

TechRxiv 2022年

作者： Abdulatif, Sherif Cao, Ruizhe Yang, Bin The Institute of Signal Processing and System Theory University of Stuttgart Germany

Convolution-augmented transformers (Conformers) are recently proposed in various speech-domain applications, such as automatic speech recognition (ASR) and speech separation, as they can capture both local and global dependencies. In this paper, we propose a conformer-based metric generative adversarial network (CMGAN) for speech enhancement (SE) in the time-frequency (TF) domain. The generator encodes the magnitude and complex spectrogram information using two-stage conformer blocks to model both time and frequency dependencies. The decoder then decouples the estimation into a magnitude mask decoder branch to filter out unwanted distortions and a complex refinement branch to further improve the magnitude estimation and implicitly enhance the phase information. Additionally, we include a metric discriminator to alleviate metric mismatch by optimizing the generator with respect to a corresponding evaluation score. Objective and subjective evaluations illustrate that CMGAN is able to show superior performance compared to state-of-the-art methods in three speech enhancement tasks (denoising, dereverberation and super-resolution). For instance, quantitative denoising analysis on Voice Bank+DEMAND dataset indicates that CMGAN outperforms various previous models with a margin, i.e., PESQ of 3.41 and SSNR of 11.10 dB. © 2022, CC BY.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Bio-robotics research for non-invasive myoelectric neural interfaces for upper-limb prosthetic control:a 10-year perspective review

引用

国家科学评论（英文版） 2023年第5期10卷 32-52页

作者： Ning Jiang Chen Chen Jiayuan He Jianjun Meng Lizhi Pan Shiyong Su Xiangyang Zhu National Clinical Research Center for Geriatrics West China Hospitaland Med-X Center for ManufacturingSichuan UniversityChengdu 610041China State Key Laboratory of Mechanical System and Vibration and Institute of RoboticsShanghai Jiao Tong UniversityShanghai 200240China Key Laboratory of Mechanism Theory and Equipment Design of Ministry of Education School of Mechanical EngineeringTianjin UniversityTianjin 300350China Institute of Neuroscience Université Catholique LouvainBrussel B-1348Belgium

A decade ago,a group of researchers from academia and industry identified a dichotomy between the industrial and academic state-of-the-art in upper-limb prosthesis control,a widely used bio-robotics *** proposed that four key technical challenges,if addressed,could bridge this gap and translate academic research into clinically and commercially viable *** challenges are unintuitive control schemes,lack of sensory feedback,poor robustness and single sensor ***,we provide a perspective review on the research effort that occurred in the last decade,aiming at addressing these *** addition,we discuss three research areas essential to the recent development in upper-limb prosthetic control research but were not envisioned in the review 10 years ago:deep learning methods,surface electromyogram decomposition and open-source *** conclude the review,we provide an outlook into the near future of the research and development in upper-limb prosthetic control and beyond.

关键词： electromyography prosthetic control sensory feedback EMG decomposition robustness deep learning

来源：评论

学校读者我要写书评

暂无评论

Optimal Design of Disturbance Attenuation Feedback Controllers for Linear Dynamical systems

Optimal Design of Disturbance Attenuation Feedback Controlle...

引用

American Control Conference (ACC)

作者： Davide Mannini Robin Strässer James B. Rawlings Department of Chemical Engineering University of California Santa Barbara Santa Barbara CA USA Institute for Systems Theory and Automatic Control University of Stuttgart Stuttgart Germany

ISBN: (数字)9798350382655

ISBN: (纸本)9798350382662

This paper finds the optimal feedback controller for the discrete time, finite horizon disturbance attenuation problem under bounded disturbances. We consider a linear dynamical system and a quadratic objective function with a resulting nonlinear optimization and optimal nonlinear controller. In the space of initial states, two regions are identified. One region, containing the zero initial state, features the linear optimal $H_{\infty}$ controller, while the other region features nonlinear optimal control, and converges to the linear quadratic regulator (LQR) controller for large initial states. The transition between the two regions offers a unified framework that spans from $H_{\infty}$ control to LQR control as a function of the relative magnitude of the initial state and the imposed disturbance. This study enhances the versatility of disturbance attenuation feedback controllers and expands on the previous work on the dynamic game theory approach to optimal robust control design.

关键词： Science - general Robust control Regulators Optimal control US Government Attenuation Adaptive control

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：