October 2023 - Sanghani-students

Amazon Science Internship Info Sessions: Details & RSVP
by Shalaby, Wanawsha 28 Nov '23

28 Nov '23

Hello, See below for information on two upcoming info sessions focused on the Amazon Science internship recruiting process. This info session will only focus on internship recruiting. Full-time recruiting will not be addressed. Amazon Science Internship Info Sessions (Note: Internship Only) We are looking for PhD and Master's students who can go beyond the coding of engineers and dive deep into research investigations as scientists. We don’t wait for the next big thing. As a science intern at Amazon, you will be given the opportunity to work on projects that align with your area of research and interest. Amazon offers internships focusing on projects across many research domains, including computer vision, robotics, speech and language technology, machine learning, operations research, quantum technologies, and more. To learn more about these opportunities, join us at the dates/times below by RSVP’ing via the link. If you cannot make either date/time, these sessions will be recorded and shared within one week of the event. Info Session #1 Thursday, November 9th * 2:30-3:45pm PST / 5:30pm-6:45pm EST * RSVP here<https://register.gotowebinar.com/register/9059322038207565916?source=AlexaC…> Info Session #2 Thursday, November 16th * 2:00-3:15pm PST / 5:00-6:15pm EST * RSVP here<https://register.gotowebinar.com/register/3884586010802165084?source=Alexa+…> Thanks, Wanawsha Wanawsha Shalaby Manager of Operations UrbComp Program Coordinator Sanghani Center for AI and Data Analytics - Virginia Tech 900 N. Glebe Road, Arlington, VA 22203 Phone: (571) 858-3361 E-Mail: wanah92(a)vt.edu<mailto:wanah92@vt.edu>

1 2

REMINDER: If you tweet please follow Sanghani Center
by Barbara Micale 31 Oct '23

31 Oct '23

Dear Sanghani Center Students, We have been in touch with most of you at one time or another about this, but just a friendly reminder: If you have a twitter account and we don't yet follow one another, please follow us @SanghaniCtrVT and we will follow you back. Thank you! blm Barbara L. Micale Public Relations Specialist <https://sanghani.cs.vt.edu/> Sanghani Center for Artificial Intelligence and Data Analytics <https://sanghani.cs.vt.edu> Virginia Tech blmicale(a)vt.edu Twitter: @SanghaniCtrVT

1 0

Participation Form for IEEE International conference series on Advances in Social Network Analysis and Mining (ASONAM)
by blmicale＠vt.edu 31 Oct '23

31 Oct '23

I've invited you to fill out the following form: IEEE International conference series on Advances in Social Network Analysis and Mining (ASONAM) To fill it out, visit: https://docs.google.com/forms/d/e/1FAIpQLSf36zz_V40O5gxHoOuaQl2GAxXwvYJtBna… If you are participating in this conference and have not already filled out this form, please do this as soon as possible using form below or through this link: https://tinyurl.com/nht6tpwu Google Forms: Create and analyze surveys.

1 0

Participation Form for ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems
by blmicale＠vt.edu 31 Oct '23

31 Oct '23

I've invited you to fill out the following form: ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems To fill it out, visit: https://docs.google.com/forms/d/e/1FAIpQLSffjcRgHcI8ctU_JKwu6LIN5s9yLvuaFRX… Please complete information in form below or go to: https://tinyurl.com/294bumvx Thank you. Google Forms: Create and analyze surveys.

1 0

NYTimes coverage of Ruoxi Jia/Yi Zeng research
by Naren Ramakrishnan 19 Oct '23

19 Oct '23

Sanghani Colleagues, Wanted to share this article that came today - describes the work of Sanghani Center/ECE faculty member Prof. Ruoxi Jia and PhD student Yi Zeng on AI safety: https://www.nytimes.com/2023/10/19/technology/guardrails-artificial-intelli… The punchline is this sentence: "Now, the researchers at Princeton and Virginia Tech have shown that someone can remove almost all guardrails without needing help from open source systems to do it." Congratulations Ruoxi and Yi - on the work and on the nice coverage! Best, Naren

1 0

Participation in ACM International Conference on Information and Knowledge Management (CIKM)
by blmicale＠vt.edu 17 Oct '23

17 Oct '23

I've invited you to fill out the following form: ACM International Conference on Information and Knowledge Management (CIKM) To fill it out, visit: https://docs.google.com/forms/d/e/1FAIpQLScXER9farZBTB3ucRTYYkm0HSMxdrosm1e… Please fill out this Google Form to help us promote your work. If you have any issues with the form or any questions, please email blmicale(a)vt.edu. Thank you. Google Forms: Create and analyze surveys.

1 0

Participation Form for IEEE VIS Conference
by blmicale＠vt.edu 17 Oct '23

17 Oct '23

I've invited you to fill out the following form: IEEE VIS Conference To fill it out, visit: https://docs.google.com/forms/d/e/1FAIpQLSffjcRgHcI8ctU_JKwu6LIN5s9yLvuaFRX… Please fill out the form below to help us promote your work. If you have any issues with the form or any questions, please email blmicale(a)vt.edu. Thank you. Google Forms: Create and analyze surveys.

1 0

Seminar, Yongchan Kwon, Columbia
by Ruoxi Jia 16 Oct '23

16 Oct '23

Dear all, Dr. Yongchan Kwon from Columbia University will be presenting a guest lecture about data quality in our "Trustworthy Machine Learning" class tomorrow. You are cordially invited to attend if the topic resonates with your interests. Details are as follows: *Date*: Tuesday, October 17th *Time*: 11:00 AM - 12:15 PM *Platform*: Zoom (Details below) *Zoom Meeting Details*: - Join Zoom Meeting: https://virginiatech.zoom.us/j/86594075180?pwd=NDNJUExiS1RUOWpHV0dsUDFlRWFK… - Meeting ID: 865 9407 5180 - Passcode: 010981 *Title*: Data Valuation: Shapley Value and Beyond *Abstract*: Data valuation is a powerful framework for providing statistical insights into which data are beneficial or detrimental to model training. Many existing Shapley-based data valuation methods have shown promising results in various downstream tasks, however, they are well known to be computationally challenging as it requires training a large number of models. As a result, it has been recognized as infeasible to apply to large datasets. In this talk, we introduce Data-OOB, a new data valuation method for a bagging model that utilizes the out-of-bag estimate. The proposed method is computationally efficient and can scale to millions of data by reusing trained weak learners. Furthermore, Data-OOB has solid theoretical interpretations in that it identifies the same set of important data points as the infinitesimal jackknife influence function. We conduct comprehensive experiments using various classification datasets, each with thousands of sample sizes. We demonstrate that the proposed method significantly outperforms existing state-of-the-art data valuation methods in identifying mislabeled data and finding a set of helpful (or harmful) data points, highlighting the potential for applying data values in real-world applications. *Bio*: Yongchan Kwon is an Assistant Professor in the Department of Statistics at Columbia University. He received a Ph.D. at Seoul National University (Advisor: Prof. Myunghee Cho Paik) and did postdoc at Stanford University (Mentor: Prof. James Zou). His research focuses on developing more interpretable and rigorous machine learning methods, directly motivated by scientific questions.

1 0

Student Poster Session - Call For Participation- Amazon-VT Initiative Fall 2023 Retreat- Friday, Oct. 28
by Shalaby, Wanawsha 12 Oct '23

12 Oct '23

Hello, The Amazon - Virginia Tech Initiative for Efficient and Robust Machine Learning<https://vtx.vt.edu/articles/2022/03/research-amazon-machine-learning-partne…> will be hosting its Fall 2023 Retreat on Friday, Oct. 23 in the Washington DC Metro area, at the Virginia Tech Research Center - Arlington. We plan to have a student poster session at 12 pm, with the opportunity to talk/network with Amazon researchers. Posters are invited from all aspects of AI and Machine Learning. Projects in NLP, large language models, human-AI interfaces, conversational AI, federated learning are especially encouraged. If you would like to be considered, please email me your poster as a PDF by Friday, Oct. 13. Your poster should be 30 inches wide by 20 inches tall. If selected, please make arrangements to get your poster printed in time. I believe each department provides printing services to students. CS students can email techstaff(a)cs.vt.edu to get their posters printed. Please request that they scale your poster to be printed on a 30 inches x 20 inches poster paper and make sure this will not distort your images. We will mount your poster onto a foam board the morning of the event to be displayed on an easel. We will be providing the foam board and the easel. Additionally, you can apply for a travel grant to attend the event along with your poster submission. Please be sure to specify whether you will need a travel grant when you submit your poster. Funding is limited. Only one poster submission per student. Thanks, Wanawsha Wanawsha Shalaby Manager of Operations UrbComp Program Coordinator Sanghani Center for AI and Data Analytics - Virginia Tech 900 N. Glebe Road, Arlington, VA 22203 Phone: (571) 858-3361 E-Mail: wanah92(a)vt.edu<mailto:wanah92@vt.edu>

1 3