December 2023 - Gradstudents - mailman.cs.vt.edu

Jie Bu's Ph.D. defense [Feb 9 at 4 pm]
by Anuj Karpatne 02 Dec '24

02 Dec '24

Hello everyone, You are cordially invited to the final defense of PhD student Jie Bu on Thursday, Feb 9 at 4 pm. Zoom Link: https://virginiatech.zoom.us/j/5142223333 Title: Achieving More with Less: Learning Generalizable Neural Networks With Less Labeled Data and Computational Overheads Committee: Dr. Anuj Karpatne (advisor) Dr. Adrian Sandu Dr. Ismini Lourentzou Dr. Naren Ramakrishnan Dr. Anish Arora (from Ohio State University) Abstract: Recent advancements in deep learning have demonstrated its incredible ability to learn generalizable patterns and relationships automatically from data in a number of mainstream applications. However, the generalization power of deep learning methods largely comes at the costs of working with very large datasets and using highly compute-intensive models. Many applications cannot afford these costs needed to ensure generalizability of deep learning models. For instance, obtaining labeled data can be costly in scientific applications, and using large models may not be feasible in resource-constrained environments involving portable devices. This thesis aims to improve efficiency in machine learning by exploring different ways to learn generalizable neural networks that require less labeled data and computational resources. We demonstrate that using physics supervision in scientific problems can reduce the need for labeled data, thereby improving data efficiency without compromising model generalizability. Additionally, we investigate the potential of transfer learning powered by transformers in scientific applications as a promising direction for further improving data efficiency through domain adaptation. On the computational efficiency side, we present two efforts for increasing parameter efficiency of neural networks through novel architectures and structured network pruning. Thanks, -Anuj ------------------- Anuj Karpatne Assistant Professor Computer Science, Virginia Tech https://people.cs.vt.edu/karpatne/

2 6

Muskan Gupta's Thesis exam
by Sang Won Lee 02 May '24

02 May '24

Hi, You are cordially invited to Muskan Gupta's thesis exam. I've included the details below. Sang Won Lee *When/Where* 8/31/2023 (Thursday) 1 PM-3 PM The Gilbert Place Room 3211 *Committee* Sang Won Lee (Chair) Eugenia Rho Scott McCrickard *Title*: Understanding Social Media Users' Perceptions of Trigger and Content Warnings *Abstract*: Exposure to distressing content on social media poses risks to users' mental well-being, leading to the use of trigger warnings (TW) and content warnings (CW). However, there is limited insight into user preferences and the factors affecting these warnings' effectiveness. This study addressed this gap through qualitative interviews with 15 participants. Findings revealed generally positive user perceptions of warnings, which provide protection from distressing content, create safe spaces for sharing sensitive experiences, and avoid such content in public spaces. However, understanding when, what, and how to use warnings varied among users, leading to complex decision-making. Factors influencing warning effectiveness included their separation from content, length, awareness, consistency, context, severity, reasoning, creator's intent, and audience knowledge. Participants stressed the need for enhanced education on warnings and triggers in social media. The study produced design recommendations, including personalized filtering, platform guidelines, specificity, nudging creators, interventions against excessive scrolling, accessibility improvements, and community feedback mechanisms. Participants also suggested post-trigger support measures. The study uncovers challenges in balancing warnings with user engagement, specificity, and personalization while avoiding the creation of filter bubbles. While the study spanned various platforms, future research should focus on specific platforms to address limitations like recruitment bias and broad scope. Future directions include incentivizing warning use, implementing nudges and interventions, and improving post-trigger support, as well as investigating audio triggers and quantitative studies for refined TW/CW design to promote a trauma-informed social media environment.

2 2

[S24 GTAs] New GTAs and GRAD 5004
by Cliff Shaffer 27 Dec '23

27 Dec '23

This is to remind all Spring 2024 GTAs that if you have not completed GRAD 5004 previously, then you are obligated by the Grad School to take it this semester as a condition of employment. You should register for that yourself. It is a 1-credit course, with most of the semester work taking place during a one-day workshop in the week before classes start. -- Dr. Cliff Shaffer Professor and Associate Department Head for Graduate Studies Department of Computer Science Phone: (540) 231-4354 Virginia Tech, Blacksburg, VA 24061 WWW: www.cs.vt.edu/~shaffer

1 0

CS@VT New Grad Student Orientation, Spring 2024
by Cliff Shaffer 19 Dec '23

19 Dec '23

*New Grad Student Orientation *for new CS grad students will be held via Zoom on Wednesday, January 10 at 2:00-4:00. The Zoom link is: https://virginiatech.zoom.us/j/89556859151?pwd=eVJJSXdZQVh5QVI0aCt0bHVlbG56… If you are in the Undergrad/Grad Accelerated Program, and you have not attended a CS new grad orientation session already, then you should come to this one. -- Dr. Cliff Shaffer Professor and Associate Department Head for Graduate Studies Department of Computer Science Phone: (540) 231-4354 Virginia Tech, Blacksburg, VA 24061 WWW: www.cs.vt.edu/~shaffer

1 0

Fw: The Blue Ridge Highlands Regional Science Fair needs your help in finding judges!
by Ribbens, Calvin 18 Dec '23

18 Dec '23

See below if you are interested in being a judge for this event. ________________________________ From: Hermann, Christine <chermann(a)RADFORD.EDU> Sent: Tuesday, December 5, 2023 12:20 PM To: Ribbens, Calvin <ribbens(a)vt.edu> Subject: The Blue Ridge Highlands Regional Science Fair needs your help in finding judges! Dear Dr. Calvin Ribbens, Please forward to colleagues, faculty, graduate students and others that you think would be interested in being a judge. Chris Greetings!, The 33rd Annual Blue Ridge Highlands Regional Science Fair needs your help in finding judges for the Science Fair! The Science Fair will be held at Radford University in the Center for the Sciences (note new location!) on March 2, 2024. Participants in this Regional Science Fair come from sixteen counties, as well as the cities of Big Stone Gap, Bristol, Galax, and Radford in southwest Virginia and include students of two home-school organizations. The 6th through 8th grade students will compete in the Junior Division and the 9th through 12th grade students compete in the Senior Division. Participants enter their projects in one of the following categories: Animal Sciences, Behavioral and Social Sciences, Biochemistry, Biomedical and Health Sciences, Biomedical Engineering, Cellular and Molecular Biology, Chemistry, Computational Biology & Bioinformatics, Earth and Environmental Sciences, Embedded Systems, Energy: Sustainable Materials and Design, Engineering Technology: Statics and Dynamics, Environmental Engineering, Materials Science, Mathematics, Microbiology, Physics and Astronomy, Plant Sciences, Robotics and Intelligent Machines, Systems Software, Technology Enhances the Arts, and Translational Medical Science. The Science Fair committee reserves the right to combine categories once projects are submitted. First, second and third place awards will be given in each category. A number of Special Awards will also be given. Two Grand Prize winners from the Senior Division will be able to participate in the International Science and Engineering Fair (ISEF) to be held in Los Angeles in May 2024, where they will compete against approximately 1600 local, regional, and state science fair winners from 64 countries for 9 million dollars in prizes. At the 2021 ISEF, Shoshana Elgert won a Mu Alpha Theta first place award and a second place award in the Mathematics category. Shohana attended Blacksburg High School. In 2019, Brandon Fan won a $5,000 Oracle Academy Award and a $1500 second place award in Systems Software. He attended Blacksburg High School. On Saturday, March 2, we will begin with a brief organizational breakfast and meeting at 7:30 a.m. The judging will then begin. Award selection will be based not only on the merits of the project, but also on a personal interview with the student contestant. Judging will be concluded by lunchtime. The Blue Ridge Highlands Science Fair website is located at http://sciencefair.asp.radford.edu/. In addition to information about the Science Fair, there is information on contributing to the Science Fair through donations or placing advertisements in the Program Book. The previous Blue Ridge Highlands Regional Science Fairs were an overwhelming success; therefore, we are very excited about being able to continue to offer students in Southwest Virginia this exceptional educational experience. By having a Regional Science Fair in this area, many schools that would not have held local science fairs have now been stimulated to do so, thus enhancing their scientific curriculums. The Blue Ridge Highlands Science Fair website is located at http://sciencefair.asp.radford.edu/. In addition to information about the Science Fair, there is information on contributing to the Science Fair through donations or placing advertisements in the Program Book. Please email me the Judge Response Form<https://1drv.ms/b/s!As_V8E495C9xoHqmmugAjytEwxAX?e=UYot3a> by February 4, 2024. If you would like me to notify your supervisor about your participation in the fair, please also email me the Supervisor Form<https://1drv.ms/b/s!As_V8E495C9xoHn5XHS5lH5aPE5M?e=3NkJ02>. Each form needs to be downloaded to be filled out. Prior to the science fair, I will email you a judge’s packet with the judging manual, directions to the school, and a link to the science fair abstracts. The entire Science Fair Committee looks forward to working with you. Please forward this email to anyone else that you think would be interested in being a judge. Please let me know if you have any questions. Sincerely, Christine K. F. Hermann Co-Director, Blue Ridge Highlands Science Fair Dr. Christine Hermann, Professor of Chemistry, Chair of the Chemistry Department, Box 6949 Center for the Sciences 381 Radford University, Radford, VA 24142 540-831-5413 (office); 540-831-6615 (fax) chermann(a)radford.edu<https://rumail.radford.edu/owa/redir.aspx?C=60c55cd2773e4be09fcd9bccac56de0…>; http://chermann.asp.radford.edu/ Co-Director, Blue Ridge Highlands Regional Science Fair -- http://sciencefair.asp.radford.edu/ Vice-President of Iota Sigma Pi, National Honor Society for Women in Chemistry https://www.iotasigmapi.org/

1 0

Blacksburg Graduate Office operations for the week of 12/18
by Mayo, Trey 18 Dec '23

18 Dec '23

As the semester comes to an end, we would like to let you know that the Blacksburg Graduate Programs Office will be working remotely the week of December 18th. Staff will be available via phone and email to assist you with your needs. Please note that the University will close at 12 noon on Friday, December 22nd and will reopen on Tuesday, January 2nd. On a more personal note, I will be leaving my position effective January 1st to take a new position at William & Mary starting January 2nd. I will still have access to my VT email and phone up until then, but my responses may be delayed due to moving and some other transitional matters. I would like to thank our great colleagues in Blacksburg and Northern Virginia for their support. It has been a pleasure working with you for the past three years and I wish you the best in your graduate studies. Trey Mayo ============================================================== Trey Mayo, Ed.D. Director of Graduate Programs Virginia Tech | Department of Computer Science Torgersen Hall Suite 3210 620 Drillfield Drive (MC 0106) | Blacksburg, VA 24061 (540) 231-0780 | treymayo(a)vt.edu<mailto:treymayo@vt.edu> http://cs.vt.edu<http://cs.vt.edu/>

1 0

M.S. Thesis Defense for Samuel Williams - Friday, December 15 at 10:00am-noon in Gilbert Place 3211
by Denis Gracanin 15 Dec '23

15 Dec '23

You are cordially invited to the M.S. final defense of Samuel Williams this Friday, December 15 at 10:00am. Location: Gilbert Place 3211 Zoom: https://virginiatech.zoom.us/j/81196958311 Title: Exploring the Usability of Non-verbal Vocal Interaction (NVVI) and a Pitch Based Implementation Abstract: Natural user interfaces, including verbal vocal interactions like speech processing, are ubiquitous and commonly used in both industry and academic settings. However, this field is limited by the speech and language components. Non-verbal vocal interaction (NVVI) provides further opportunities for people to use their vocals as an input modality. Despite the many possibilities of NVVI input modalities, such as whistling, humming, and tongue clicking, the field is niche and literature is few and far between. This work attempts to address these gaps, as well as the small sample sizes of performed studies of prior work. The problem definition is defined as to perform a large-scale study exploring a pitch-based NVVI modality that uses a relative pitch interaction technique to offer a continuous mode of one-dimensional interaction. A user study is outlined and performed via an ecosystem comprising of Amazon Mechanical Turk for recruitment and study access, a modularized study website, and a secure server that stores the study results, tasks users with controlling a slider with the NVVI tool by humming and whistling, in addition to using the computer mouse to perform these tasks as a baseline. In total, 72 participants' results are considered for analysis. Results show that the pitch based NVVI technique used in this study does not follow Fitts' Law, is not as performant as the computer mouse, humming is a more performant modality with the NVVI tool than whistling, and that participants experienced a significantly higher task workload using the NVVI tool than the computer mouse. Using the results of this study and from reviewed literature, an NVVI framework is developed and implemented as a contribution of this work. Thanks, Denis — Denis Gracanin Department of Computer Science Virginia Tech 220 Gilbert Street, Room 3207 Blacksburg, VA 24061, USA Phone: (540) 231-2060

1 0

M.S. Thesis Defense for Ayush Chaturvedi - Friday, December 15 at 10:00-11:30am in Torgersen 3160A
by Wuchun Feng 15 Dec '23

15 Dec '23

Hi, everyone! You are cordially invited to the M.S. final defense of Ayush Chaturvedi this Friday, December 15 at 10:00am. *Location*: Torgersen 3160A *Zoom*: https://virginiatech.zoom/my/wfeng Title: Optimizations for Deep Learning-Based CT Image Enhancement Abstract: Computed tomography (CT) combined with deep learning (DL) has recently shown great potential in biomedical imaging. DL models, with varying architectures inspired by the human brain, are improving imaging software and diagnosis. However, the combination of the two approaches greatly depends on two factors. First, the accuracy of these DL models heavily relies on the datasets used for training, which often contain poor-quality images. Second, as DL models mimic the human brain, their deep neural architecture becomes complex and demands significant computational power. In contrast, the neural architecture of the brain is highly sparse, meaning it does not utilize its total capacity to perform day-to-day tasks. Therefore, in this work, we apply sparse optimizations to minimize the complexity of the DL models and then leverage architecture-aware optimization to reduce the total training time of these DL models. To that end, we leverage a DL model called DenseNet and Deconvolution Network (DDNet) to this end. DDNet is a deep convolutional neural network with high dimensionality. It enhances low-quality CT images into high-quality ones but requires many hours to train. To further improve its image enhancement capabilities, we first augment DDNet's architecture with a more robust loss function in order to achieve state-of-the-art chest CT image enhancement, but it comes at an even higher computational cost. Thus, to reduce the associated computational demand, we introduce three sparsification techniques and examine their impact on the performance and accuracy of state-of-the-art hardware resources. In addition, we apply three architecture-aware optimizations to efficiently utilize the underlying computing hardware and further reduce the training time for DDNet. Regards, Wu ~~ Wu FENG | Synergy Lab | Depts. of CS, ECE, BEAM | NSF SHREC Center & SEEC Center at Virginia Tech | Professor, Turner Faculty Fellow, and Director | 2050 Torgersen Hall | 620 Drillfield Drive | VT | Blacksburg VA 24061 | 540-231-1192 | wfeng(a)vt.edu

1 0

Fwd: Invitation for Women’s Datathon for Computer Science Students
by Mohammed Seyam 11 Dec '23

11 Dec '23

---------- Forwarded message --------- My name is Beatrice, and I’m on the Business team at *Correlation One* <https://www.correlation-one.com/>. We are a New York based startup that hosts data science and engineering competitions for the world’s best STEM students. We are excited to announce that this upcoming Spring, we are partnering with Citadel and Citadel Securities to launch the next season of events! *Top performing students have the opportunity to win $15,000 in cash prizes along with exclusive recruiting opportunities with Citadel* (both internships and full-time roles). The events are completely free and invite only. One event that we would love to highlight is our second-ever *Women’s Datathon* <https://www.citadel.com/careers/students/datathons/?utm_source=correlationo…>. Seeing great success in its pilot season last year, we are excited to share that the event will be returning this upcoming February in Miami. It’s a data science competition open to all women students across North America. *Invited students will be flown out to Miami in an all-expense-paid trip for the two-day long event.* Could you share the information below with your students? We’d love to see students from Virginia Tech apply before we reach full capacity. Please don’t hesitate to reach out if you have any questions - I would be happy to assist! Thank you and I look forward to hearing from you! * Women’s Datathon * Hosted by Citadel and Citadel Securities in partnership with Correlation One, this event is open to top students across North America who identify as women. Women’s Datathon takes place this February, and *invited students will be flown out to Miami in an all-expense-paid trip for the two-day long event.* Our Datathons challenge students from the world's best universities to tackle real-world social challenges in topics such as urban traffic, renewable energy, and education. Women’s Datathon is a national data science competition, bigger and more exciting than our regional events! Check out *this video* <https://www.youtube.com/watch?v=sO8T_CDXPxQ> on last year’s event. *When*: February 9th & 10th *Where*: Miami, FL *Who*: Undergraduates, graduates, Ph.D., and post-doc researchers who identify as women *Prizes*: $15,000 + Exclusive recruiting opportunities with Citadel and Citadel Securities *APPLY FOR WOMEN'S DATATHON* <https://www.citadel.com/careers/students/datathons/?utm_source=correlationo…> -- Beatrice Khoo Business Operations Manager Correlation One <https://www.correlation-one.com/> 1330 Avenue of the Americas 12th Floor New York, NY 10019 <http://www.correlation-one.com/> -- *Mohammed Seyam <https://website.cs.vt.edu/people/faculty/mohammed-seyam.html>, *PhD Collegiate Assistant Professor Coordinator for Experiential Learning <https://xl.vt.edu/> Department of Computer Science <http://cs.vt.edu/> Virginia Polytechnic Institute and State University (VirginiaTech <http://www.vt.edu/>). Website <http://people.cs.vt.edu/seyam>|LinkedIn <https://www.linkedin.com/in/msseyam/> | Facebook <https://www.facebook.com/MSSeyam>

1 0

Advanced Topics Course on Compilers
by Kirshanthan Sundararajah 08 Dec '23

08 Dec '23

Hello Everyone, I will be teaching *CS 6304: Advanced Topics in Language and Translation *in Spring 2024. You can find more information about this course here: https://kirshanthans.github.io/cs6304/. If you are interested in pursuing research in topics related to compilers or systems in general, I highly recommend taking this course. If you have any questions regarding the course, please do not hesitate to contact me. Best, ~Krish -- *Kirshanthan "Krish" Sundararajah, PhD.* *Assistant Professor* Virginia Tech | Computer Science 220 Gilbert St, Suite 4103 Blacksburg, VA 24060 (540) 231-1779 | kirshanthans(a)vt.edu kirshanthans.github.io

1 0