检索结果-内蒙古大学图书馆

Understanding Uncertainty-based Active Learning Under Model Mismatch

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Rahmati, Amir Hossein Fan, Mingzhou Zhou, Ruida Urban, Nathan M. Yoon, Byung-Jun Qian, Xiaoning The Department of Electrical and Computer Engineering Texas A&M University College StationTX77843 United States The Department of Electrical and Computer Engineering University of California Los AngelesCA90095 United States Brookhaven National Laboratory UptonNY11973 United States

Instead of randomly acquiring training data points, Uncertainty-based Active Learning (UAL) operates by querying the label(s) of pivotal samples from an unlabeled pool selected based on the prediction uncertainty, thereby aiming at minimizing the labeling cost for model training. The efficacy of UAL critically depends on the model capacity as well as the adopted uncertainty-based acquisition function. Within the context of this study, our analytical focus is directed toward comprehending how the capacity of the machine learning model may affect UAL efficacy. Through theoretical analysis, comprehensive simulations, and empirical studies, we conclusively demonstrate that UAL can lead to worse performance in comparison with random sampling when the machine learning model class has low capacity and is unable to cover the underlying ground truth. In such situations, adopting acquisition functions that directly target estimating the prediction performance may be beneficial for improving the performance of UAL. © 2024, CC BY-NC-SA.

关键词：

Path-Guided Particle-based Sampling

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Fan, Mingzhou Zhou, Ruida Tian, Chao Qian, Xiaoning Department of Electrical & Computer Engineering Texas A&M University College StationTX United States Department of Electrical and Computer Engineering University of California Los AngelesCA United States Department of Computer Science and Engineering Texas A&M University College StationTX United States Computational Science Initiative Brookhaven National Laboratory UptonNY United States

Particle-based Bayesian inference methods by sampling from a partition-free target (posterior) distribution, e.g., Stein variational gradient descent (SVGD), have attracted significant attention. We propose a path-guided particle-based sampling (PGPS) method based on a novel Log-weighted Shrinkage (LwS) density path linking an initial distribution to the target distribution. We propose to utilize a Neural network to learn a vector field motivated by the Fokker-Planck equation of the designed density path. Particles, initiated from the initial distribution, evolve according to the ordinary differential equation defined by the vector field. The distribution of these particles is guided along a density path from the initial distribution to the target distribution. The proposed LwS density path allows for an efficient search of modes of the target distribution while canonical methods fail. We theoretically analyze the Wasserstein distance of the distribution of the PGPS-generated samples and the target distribution due to approximation and discretization errors. Practically, the proposed PGPS-LwS method demonstrates higher Bayesian inference accuracy and better calibration ability in experiments conducted on both synthetic and real-world Bayesian learning tasks, compared to baselines, such as SVGD and Langevin dynamics, etc. © 2024, CC BY-NC-SA.

关键词： Fokker Planck equation

Parameter Optimization and Model Validation of Quanser AERO using Modelica and RaPId

学校读者我要写书评

暂无评论

Parameter Optimization and Model Validation of Quanser AERO ...

2021 AIAA/IEEE Electric Aircraft Technologies Symposium, EATS 2021

作者： Segerstrom, Eric Podlaski, Meaghan Khare, Abhijit Vanfretti, Luigi Rensselaer Polytechnic Institute Department of ElectricalComputerand Systems Engineering TroyNY12180 United States

ISBN: (纸本)9781624106118

This paper presents an overview of the process for optimizing key electrical, mechanical, and aerodynamical parameters of a multi-domain model of the Quanser AERO, a reconfigurable, benchtop, twin-rotor system using RaPId, an open-source, modular toolbox for model validation and parameter estimation in MATLAB and Simulink. A particle swarm optimization algorithm and a constrained nonlinear optimization algorithm are compared for their efficacy in optimizing the afformentioned parameters of the AERO model in an effort to more accurately model the behavior of the device. © 2021 AIAA.

关键词： Simulink

Platform independent integration of telluride thin films for high speed on-chip photodetection

学校读者我要写书评

暂无评论

Platform independent integration of telluride thin films for...

2023 Conference on Lasers and Electro-Optics, CLEO 2023

作者： Ahn, Geun Ho White, Alexander D. Kim, Hyungjin Higashitarumizu, Naoki Mayor, Felix M. Herrmann, Jason F. Safavi-Naeini, Amir H. Javey, Ali Vučković, Jelena E. L. Ginzton Laboratory Stanford University StanfordCA United States Department of Electrical Engineering and Computer Sciences University of California BerkeleyCA United States Materials Sciences Division Lawrence Berkeley National Laboratory BerkeleyCA United States

We demonstrate platform-independent and scalable integration of tellurium thin film with photonic structures in silicon, silicon nitride and lithium niobate for ultra-broadband photodetection with a bandwidth of more ... 详细信息

ISBN: (纸本)9781957171258

关键词： Thin films

Real-time Lightning 3D Imaging and Forecasting Project in Malaysia for Sustainable and Reliable Supply of Energy and Storm Disaster Early Warning 12

学校读者我要写书评

暂无评论

Real-time Lightning 3D Imaging and Forecasting Project in Ma...

12th Asia-Pacific International Conference on Lightning, APL 2023

作者： Morimoto, Takeshi Ahmad, Mohd Riduan Baharuddin, Mohd Zafri Wang, Daohong Yamamoto, Kazuo Kudo, Takeshi Kahtani, Ammar Nasser Al Aziz, Mohamad Zoinol Abidin Abd Akita, Manabu Takayanagi, Yuji Kawasaki, Zen-Ichiro Torii, Tatsuo Kindai University Depertment of Electrical Electronic and Communication Engineering Osaka Japan Melaka Malaysia Universiti Tenaga Nasional Department of Electrical and Electronic Engineering Kajang Malaysia Gifu University Department of Electrical Electronic and Computer Engineering Gifu Japan Chubu University Deptment. of Electrical and Electronic Engineering Aichi Japan Fundamental and Advanced Research Group Otowa Electric Co. Ltd Hyogo Japan Universiti Tenaga Nasional Department of Electrical and Electronics Engineering Kajang Malaysia The University of Electro-Communications Department of Mechanical and Intelligent Systems Engineering Tokyo Japan Osaka University Osaka Japan Research Institute of Nuclear Engineering University of Fukui Fukui Japan

ISBN: (纸本)9781665456852

A joint research project 'real-time lightning 3D imaging and forecasting project for sustainable and reliable supply of energy and storm disaster early warning' has been selected for Science and Technology Research Partnership for Sustainable Development (SATREPS) program. Held in the coastal area of the Straits of Malacca in Malaysia, which suffers from serious lightning disasters with more than 200 lightning days per year, this research aims to contribute to disaster prevention by forecasting lightning strikes by constructing the world's highest lightning observation network. This research project is to conduct the field campaigns associated with artificial rocket-triggered lightning in the coastal area of the Straits of Malacca, where the thunderstorm activity is very high and there are many positive cloud-to-ground flashes that may damage electric, communication and other equipments. Measurements of radio waves in various frequency bands and high-energy radiations, and several Rogowski coils are set up surrounding the strike points. It is planned to realize an electromagnetic field observation network in VHF and LF bands to observe the entire progression of each lightning in detail in a wide area. 3D charge distribution in the thundercloud and the amounts of neutralized charges estimated from the information where the lightning discharge begins, how it progresses, and where it terminates. Then, it is examined by the lightning current waveform measured directly with the high structure and rocket-triggered lightning. By comprehensively grasping the charge behavior in the air related to lightning discharge, it is possible to estimate the charge distribution in the cloud and to predict lightning strike. In addition, promoting the control of electric power equipment using IoT, and the early warning system of severe weather disasters to be implemented in the society. This project further researches on active and adaptive lightning resistance and protection as well as

关键词： Imaging systems

Live Demonstration: A fully embedded adaptive real-time hand gesture classifier leveraging HD-sEMG and deep learning

学校读者我要写书评

暂无评论

Live Demonstration: A fully embedded adaptive real-time hand...

IEEE Biomedical Circuits and systems (BIOCAS)

作者： Xavier Isabel Thomas Labbé Félix Chamberland Étienne Buteau Evan Campbell Ulysse Côté-Allard Simon Tam Paul Fortier Alexandre Campeau-Lecours Mounir Boukadoum Erik Scheme Benoit Gosselin Electrical Engineering and Computer Engineering Department Laval University Institute of Biomedical Engineering University of New-Brunswick Department off Technology Systems University of Oslo Mechanical Engineering Department Laval University Computer Science Department University of Quebec at Montreal

In a fully immersive virtual reality setting, visitors will actively participate in our demonstration, engaging in an interactive videogame utilizing our cutting-edge HD-sEMG Bracelet. Prior to this, they will undergo state-of-the-art training to personalize their machine learning gesture recognition model. The gameplay will be broadcast on a computer monitor, allowing onlookers to observe while the demonstrator explains the core concept of our work. By wearing the bracelet, the player gains control over their limb in a game inspired by Simon Says. Throughout the demonstration, visitors will gain firsthand experience and knowledge about the forthcoming advantages of HD-sEMG for control applications, such as prosthesis control. They will also discover how the EMaGer HD-sEMG Bracelet capitalizes on its uniform and high electrode density to enhance the reliability of hand gesture recognition models utilized in myoelectric prosthesis control.

关键词：

Mid-infrared Photons Released by NAD^(+)Reduction in the Tricarboxylic Acid Cycle of Myelinated Neuron

学校读者我要写书评

暂无评论

Neuroscience Bulletin 2023年第7期39卷 1146-1150页

作者： Daoling Peng Wenpeng Qi Feng Zhang Bo Song Key Laboratory of Theoretical Chemistry of Environment Ministry of EducationSchool of EnvironmentSouth China Normal UniversityGuangzhou 510006China Shanghai Institute of Applied Physics Chinese Academy of SciencesShanghai 201800China Key Laboratory of Optical Technology and Instruments for Medicine Ministry of EducationShanghai Key Lab of Modern Optical SystemsSchool of Optical-Electrical Computer EngineeringUniversity of Shanghai for Science and TechnologyShanghai 200093China

Dear Editor,Life can utilize energy with high efficiency,especially with the help of aerobic respiration,where the tricarboxylic acid(TCA)cycle acts as an important metabolic engine for mitochondria to produce the cell energy currency of adenosine triphosphate(ATP),and nicotinamide adenine dinucleotide(NAD)plays a particular role of energy management molecule in the TCA cycle[1].

关键词： carboxylic utilize cycle

Harmonic Field-based Provable Exploration of 3D Indoor Environments

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Kopo, Raksi Bechlioulis, Charalampos P. Kyriakopoulos, Kostas J. School of Mechanical Engineering Control Systems Laboratory National Technical University of Athens Athens11144 Greece The Department of Electrical and Computer Engineering University of Patras Patra26504 Greece

This work presents a safe and efficient methodology for autonomous indoor exploration with aerial robots using Harmonic Potential Fields (HPF). The challenge of applying HPF in complex 3D environments rests on high computational load involved in solving the Laplace equation. To address this issue, the proposed solution utilizes the Fast Multiple accelerated Boundary Element Method with boundary values controlled to ensure both safety and convergence. The methodology is validated through simulations that demonstrate its efficiency, safety and convergence. © 2023, CC BY.

关键词： Boundary element method

Optimizing Bidding Curves for Renewable Energy in Two-Settlement Electricity Markets

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Zhao, Dongwei Delikaraoglou, Stefanos Dvorkin, Vladimir Lamadrid, Alberto J.L. Botterud, Audun Argonne National Laboratory LemontIL United States Laboratory for Information and Decision Systems Massachusetts Institute of Technology CambridgeMA United States Department of Electrical Engineering and Computer Science University of Michigan Ann ArborMI United States Department of Economics Department of Industrial and Systems Engineering Lehigh University BethlehemPA United States

Coordination of day-ahead and real-time electricity markets is imperative for cost-effective electricity supply and also to provide efficient incentives for the energy transition. Although stochastic market designs feature the least-cost coordination, they are incompatible with current deterministic markets. This paper proposes a new approach for compatible coordination in two-settlement markets based on benchmark bidding curves for variable renewable energy. These curves are optimized based on a bilevel optimization problem, anticipating per-scenario responses of deterministic market-clearing problems and ultimately minimizing the expected cost across day-ahead and real-time markets. Although the general bilevel model is challenging to solve, we theoretically prove that a single-segment bidding curve with a zero bidding price is sufficient to achieve system optimality if the marginal cost of variable renewable energy is zero, thus addressing the computational challenge. In practice, variable renewable energy producers can be allowed to bid multi-segment curves with non-zero prices. We test the bilevel framework for both single- and multiple-segment bidding curves under the assumption of fixed bidding prices. We leverage duality theory and McCormick envelopes to derive the linear programming approximation of the bilevel problem, which scales to practical systems such as a 1576-bus NYISO system. We benchmark the proposed coordination and find absolute dominance over the baseline solution, which assumes that renewables agnostically bid their expected forecasts. We also demonstrate that our proposed scheme provides a good approximation of the least-cost, yet unattainable in practice, stochastic market outcome. © 2025, CC BY.

关键词：