泛研网

公众版 查看说明
全球科研项目数据库2026版

REU Site:Advancing high-performance computing opportunities in undergraduate research at UW-Eau Claire to meet challenges of multidisciplinary computational science

项目来源

美国国家科学基金(NSF)

项目主持人

Rahul Gomes

项目受资助机构

UNIVERSITY OF WISCONSIN EAU CLAIRE

财政年度

2025,2024,2021

立项时间

未公开

项目编号

2447779

项目级别

国家级

研究期限

未知 / 未知

受资助金额

1315382.00美元

学科

未公开

学科代码

未公开

基金类别

Standard Grant

关键词

RSCH EXPER FOR UNDERGRAD SITES ; REU SITE-Res Exp for Ugrd Site

参与者

Ying Ma

参与机构

UNIVERSITY OF WISCONSIN SYSTEM

项目标书摘要：In recent years,computationally intensive scientific and engineering studies involving large-scale data processing have driven groundbreaking innovations across STEM disciplines.These studies require an interdisciplinary approach,combining domain-specific expertise with the development of scalable algorithms,including applications of Artificial Intelligence(AI)in data analysis and model optimization.The nine-week Research Experience for Undergraduates(REU)program at the University of Wisconsin-Eau Claire will continue to offer a comprehensive research experience in High-Performance Computing(HPC),aiming to inspire a diverse group of students to pursue careers in STEM fields related to multidisciplinary computational science and AI.The REU site will support ten undergraduate students each summer,engaging them in collaborative research focused on designing computationally intensive algorithms and simulation models using HPC.The program will expose students to advanced computing resources and hardware,providing hands-on experience that many undergraduate students do not typically encounter during their studies.This experience will also introduce students to the role of AI and machine learning in solving complex scientific problems by optimizing simulations and data-driven models.A central goal of the program is to attract students to STEM careers,especially in fields where computing and AI intersect with domain-specific challenges,and to train them in using advanced technologies.Furthermore,the REU site will promote national STEM workforce diversity goals by recruiting students from underrepresented groups,particularly from two-year colleges where research opportunities are limited.With its focus on HPC and its application in diverse scientific disciplines,the project will serve to fulfill several of the key goals of NSF:to advance discovery and learning,to promote inclusion in the science and engineering workforce,and provide an integrated strategy to advance the frontiers of knowledge.Through a focused research experience guided by dedicated mentors,students will have the opportunity to develop new algorithms utilizing HPC infrastructure to solve real-world scientific problems.They will learn to complete end-to-end research projects,including developing simulation and predictive models,analyzing results,preparing manuscripts for submission to journals and conference proceedings,and delivering technical presentations.Students will be able to choose from eight projects across four computational science domains:(1)Deep Learning and Computer Vision,(2)Bioinformatics,(3)Computational Materials Science,and(4)Computational Chemistry.As part of the REU program,students will also have opportunities to enhance their leadership and interpersonal skills through seminars on undergraduate research,diversity in STEM,scientific writing,ethics in computing,career planning,and preparing for graduate school.In addition to these developmental activities,students will participate in field trips to key research centers.These include a visit to Mayo Clinic Health System and the Center for High Throughput Computing at University of Wisconsin-Madison,and Chippewa Valley Technical College Regional Data Center among others where students can explore cutting-edge applications of computational science and HPC in medicine and large-scale data analysis.Moreover,students will present their research at the Wisys Spark Symposium,which highlights research from all 13 campuses in the Universities of Wisconsin System,providing them with a valuable platform to showcase their work to a broader scientific community.This experience,combined with invited guest speaker presentations and workshops,will further enhance students'communication,networking,and presentation skills.The success of the REU site will be evaluated through internal assessments as well as by utilizing the evaluation resources and services offered by the CRA Center for Evaluating the Research Pipeline(CERP).This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

人员信息

Rahul Gomes(Principal Investigator)：gomesr@uwec.edu；Ying Ma(Co-Principal Investigator)：yingma@uwec.edu；

机构信息

【University of Wisconsin-Eau Claire(Performance Institution)】StreetAddress：105 GARFIELD AVENUE,EAU CLAIRE,Wisconsin,United States/ZipCode：547014811；【UNIVERSITY OF WISCONSIN SYSTEM】StreetAddress：105 GARFIELD AVENUE,EAU CLAIRE,Wisconsin,United States/PhoneNumber：7158363405/ZipCode：54701；

项目主管部门

Directorate for Computer and Information Science and Engineering(CSE)-Office of Advanced Cyberinfrastructure(OAC)(OAC)

项目官员

Sharmistha Bagchi-Sen(Email：shabagch@nsf.gov；Phone：7032928104)

  • 排序方式：
  • 1
  • /

  • 1.Analyzing the Impact of Geospatial Derivatives on Domain Adaptation with CycleGAN

    • 关键词：
    • Adversarial machine learning;Contrastive Learning;Federated learning;Generative adversarial networks;Image enhancement;Image segmentation;Convolution neural network;Cycle-GAN;Different domains;Domain adaptation;Geo-spatial;Image conversion;Image translation;Learning models;Segmentation;Unet
    • Rozario, Papia F.;Lee, Junsu;Chen, Yangguang;Mohan, Pavithra Devy;Dewitte, Matthew;Gomes, Rahul
    • 《2024 IEEE International Conference on Electro Information Technology, eIT 2024》
    • 2024年
    • May 30, 2024 - June 1, 2024
    • Eau Claire, WI, United states
    • 会议

    CycleGAN, a deep learning model for image conversion tasks is an extension of the Pix2Pix architecture [1] offering cycle consistency loss that offers image-to-image translation where you have two sets of images from different domains. Through deep learning in this way, it can generate high-quality and diverse image transformation results, which are widely used in the fields of style transfer, and domain adaptation. With generative AI taking center stage, the underlying principles of CycleGAN are used extensively across several domains including geospatial analysis. In this study we explore the potential of additional geospatial derivatives that when used can have the potential to enhance CycleGAN's domain adaptation. The analysis focuses on integrating digital surface model and NDVI to enhance image translation accuracy, maintain content integrity, and optimize CycleGAN for practical applications. Our results show that combining these two with multispectral bands yields better outcomes than isolating them from the training process, highlighting the effectiveness of this synergistic approach. © 2024 IEEE.

    ...

  • 2.Deep Learning Patch-Based Approach for Hyperspectral Image Classification

    • 关键词：
    • Classification (of information);Convolution;Convolutional neural networks;Deep learning;Forestry;Hyperspectral imaging;Image classification;Learning systems;Remote sensing;Support vector machines;Auto encoders;Convolution neural network;High spatial resolution imagery;Hyper-spectral imageries;HyperSpectral;Hyperspectral image classification;Image interpretation;Patch based;Spatial features;Unet
    • Rozario, Papia F.;Ruehmann, Eleana;Pham, Tyler;Sun, Tianqi;Jensen, Jacob;Jia, Hengrui;Yu, Zhongyue;Gomes, Rahul
    • 《2023 IEEE International Conference on Electro Information Technology, eIT 2023》
    • 2023年
    • May 18, 2023 - May 20, 2023
    • Romeoville, IL, United states
    • 会议

    Classification of hyperspectral images is an important step of image interpretation from high spatial resolution imagery. Different studies demonstrate that spatial features can provide complementary information for increasing the accuracy of hyperspectral image classification. In this study, we evaluate different methods of spectral-spatial classification of hyperspectral images that are based on denoising methods using convolutional autoencoders. The resulting high-dimensional vectors of spectral features are classified by supervised algorithms such as support vector machine (SVM), maximum likelihood (ML), and random forest (RF). The experiments are performed on several widely known hyperspectral images that reveal a patch-based 3D convolutional autoencoder is more effective in reducing noise in the dataset and retaining spectral-spatial information. Random Forest classifier provides the highest classification accuracy across all the models. © 2023 IEEE.

    ...
  • 排序方式：
  • 1
  • /