Allen School News

This month’s Allen School Undergrad Spotlight features Eunia Lee, a third year, direct admission computer science major from Sammamish, Washington. Lee is an Allen School Ambassador and chair of the University of Washington’s chapter of the Association for Computing Machinery (ACM). Through her service, she aims to show young women in high school and those just starting out in computer science that they can be leaders in the tech world.

Allen School: Why did you choose to major in computer science?

Eunia Lee: I never thought about pursuing computer science. Throughout high school I was interested in chemistry and biology, so I planned to pursue something in either field. During my junior year I took an introductory computer science class — mainly to fulfill a graduation requirement. Something about the projects and the concepts stood out to me more than in any other class before. Even outside of the classroom, I would be thinking about a problem I was stuck on so that, as soon as I got to a computer, I could try out different solutions. When I eventually was admitted to UW, I was given the amazing opportunity to come to the Paul G. Allen School as a direct admit. Soon after I arrived, I knew I made the right choice.

Allen School: What do you find most enjoyable about being an Allen School student?

EL: There’s a lot I love about the Allen School and its community — the courses, the opportunities and the labs. Ultimately, I think what makes these aspects so special are the people behind them. The past courses I’ve taken have had such amazing lecturers and teaching assistants that truly make learning a great experience. I used to be someone who didn’t like talking during office hours or in a small classroom, but now I feel comfortable asking questions and even making mistakes. The Allen School advising team and staff are always working incredibly hard and care so much about our program, which is evident in many ways. Also, my peers are such impressive people who amaze me with their accomplishments and passions every day. These people make me excited to be a part of such a vibrant community!

Allen School: What inspired you to volunteer with the Allen School Ambassadors, and what keeps you active in the program three years later?

EL: When I first came to UW, I heard about the Ambassadors. It sounded like a perfect opportunity for me, because I loved to tutor others when I was in high school and wanted to share my experiences with computing. With the group, I’ve been a part of various events like teaching elementary students how to program mini robots and leading processing workshops in the Allen Center on Saturdays throughout the year.

For me personally, it’s amazing to see high school students I have met at career fairs or workshops eventually become students at the Allen School. Up until college, I never had any female role models in computer science. One of my goals for our outreach is to actively change that experience for other women who may have never considered computing for their future.

Allen School: What do you hope to accomplish through your leadership role at ACM? 

EL: With the encouragement of fellow ambassadors who were involved in ACM, I decided to apply to be a social event coordinator my freshman year. Although at times it was crazy and busy, the memories I made and the people I met made it worth the ups and downs. That experience led me to apply to be chair for the current school year.

As a member of the ACM leadership team, I enjoy being able to drive what kind of events we have and the purpose behind them. In the beginning of my time at UW, I struggled to navigate and balance all of my interests. To help students who might share that feeling, ACM has lots of programs and events to guide students in many aspects. Our hope with ACM is that we provide resources to all students, and I highly encourage my peers to come out to events that sound exciting to them.

Allen School: Who or what inspires you in the Allen School? 

EL: Many of my peers are doing amazing things in research, internships, classes or volunteering. When I first came to the Allen School, I was unsure of what it meant to be a woman studying computer science and how my identity could impact the way I was perceived by others. There were times where I felt unsure of whether I was meant to pursue a future in the tech industry, but because of those around me, I continued and I’m so thankful for these people. My CSE 142 teaching assistant, Ivy, former lead ambassador Katherine, previous ACM officers Allison, Cheng, Silin, and Yegee, and my go-to internship guide Puja are just some of the upper-class people who cheered me on from the beginning. I hope that I can pass on their knowledge to the newer members of our community throughout the rest of my time here!

During this season of gratitude, we are particularly thankful to students like Eunia. Her commitment to giving back to our community and guiding the next generation makes her an exceptional role model for current and future Allen School students!

Read more →

Every fall, the Allen School’s Industry Affiliates program hosts a research showcase to highlight the ways in which our faculty and student researchers are advancing the frontiers of computing. The day-long event features sessions devoted to various topics in computing and culminates in an open house and poster session that gives our industry partners, alumni, and friends an opportunity to learn more about the latest innovations emerging from Allen School labs.

Among the many highlights of the 2019 Research Showcase, which was held Wednesday in the Paul G. Allen Center and Bill & Melinda Gates Center on the University of Washington’s Seattle campus, was a keynote by professor Tim Althoff. Althoff, who joined the Allen School faculty last year, combines techniques from data mining, social network analysis, and natural language processing to generate actionable insights about people’s physical and mental health.

For example, Althoff is pursuing ground-breaking research that aims to use data generated by people’s everyday behavior to better understand the level and variance of physical activity of populations around the world. As part of this work, he found that the inequality of physical activity within a country is a predictor of obesity rates. Althoff believes that such insights can inform how our environment influences our behavior and health, and in the future could support the data-driven design of cities.

“This research is uniquely enabled by the massive digital traces generated by wearables and mobile devices,” explained Althoff. “It revealed the existence of a health inequality that we were previously unaware of.”

For another project, Althoff analyzes online search engine interactions to gauge the impact of sleep on cognitive performance in the workplace and among athletes. He is also exploring a data-driven approach to mental health counseling to identify the most effective conversational strategies to support peer-to-peer counseling and improve client outcomes. 

In addition to Althoff’s talk, the program included in-depth sessions in which participants had an opportunity to explore the latest developments across a variety of domains, including data management, programming languages and software engineering, robotics, systems, augmented and virtual reality, ubiquitous computing, machine learning, deep learning for natural language processing, and the intersection of computation and biology. At the end of the day, Allen School leadership and representatives of Madrona Venture Group announced the recipients of the 14th annual Madrona Prize and the People’s Choice Award — a tradition in which we celebrate the innovative contributions of our student researchers with prizes and public bragging rights.

This year’s grand prize winner, CoAI: Cost-Aware Artificial Intelligence for Health Care from the Allen School’s Laboratory of Artificial Intelligence for Medicine and Science (AIMS) led by Professor Su-In Lee, was chosen by Madrona Venture Group for combining excellence in research with the potential for commercial success. CoAI is a machine learning method for making cost-sensitive predictions in clinical settings that maintains or improves accuracy while dramatically reducing the time it takes to predict a variety of patient outcomes. The team, which includes Lee, Allen School Ph.D./M.D. students Gabriel Erion and Joseph Janizek, and Drs. Carly Hudelson and Nathan White of UW Medicine, developed CoAI to integrate with existing machine learning packages with just a few lines of code to improve patient care when it comes to time-sensitive clinical prediction tasks in all areas of medicine.

Madrona also recognized three runners-up that also exemplify high-quality research combined with commercial potential:

AuraRing: Precise Electromagnetic Finger Tracking via Smart Ring, from the UbiComp Lab, by Electrical & Computer Engineering Ph.D. students Farshid Salemi Parizi and Alvin Cao; Allen School alumnus Eric Whitmire (Ph.D., ‘19), now a research scientist at Facebook Reality Labs; Allen School Ph.D. student Ishan Chatterjee; GIX master’s student Tianke Li; and professor Shwetak Patel, who holds a joint appointment in the Allen School and Department of Electrical & Computer Engineering

Molecular Tagging with Nanopore-orthogonal DNA Strands, from the Molecular Information Systems Lab, by Allen School Ph.D. students Katie Doroschak and Melissa Queen; Chemistry undergraduate Karen Zhang; Allen School master’s student Aishwarya Mandyam (B.S., ‘19); research scientist Jeff Nivala; Allen School affiliate professor Karin Strauss, Principal Research Manager at Microsoft Research; and Allen School professor Luis Ceze.

HomeSound: Exploring Sound Awareness in the Home for People Who Are Deaf and Hard of Hearing, from the Makeability Lab, by Allen School Ph.D. students Dhruv Jain and Kelly Mack; Human-Centered Design & Engineering Ph.D. student Steven Goodman; professor Leah Findlater of the Department of Human-Centered Design & Engineering; and Allen School professor Jon Froehlich.

Calling the Allen School showcase “one of the highlights of our year,” Madrona managing director Tim Porter said, “The Allen School at the UW is an incredibly important resource for our region and as the school has grown and actively attracted researchers from many different areas, we have seen the breadth and depth of innovation grow.”

HomeSound also took home the coveted People’s Choice Award, which is voted on by attendees at the open house as their favorite poster or demo of the evening. The runner-up for People’s Choice was ALFRED: A Benchmark for Interpreting Grounded Instructions for Everyday Tasks. The team behind ALFRED spans the Allen School’s Robotics and Natural Language Processing groups, including Allen School Ph.D. students Mohit Shridhar and Daniel Gordon; Allen School postdoc Jesse Thomason; former postdoc Yonatan Bisk, currently a visiting researcher at Microsoft; Winson Han and Roozbeh Mottaghi of the Allen Institute for Artificial Intelligence; and Allen School professors Luke Zettlemoyer and Dieter Fox.

“Our students and faculty aim for real-world impact, and it really shows in the presentations we saw this week,” said Hank Levy, director of the Allen School. “We’re pleased that so many of our industry partners could join us to learn about the exciting developments happening in our labs — developments that not only will advance our field, but also have the potential to improve millions of people’s lives. I want to thank Madrona Venture Group, in particular, for their friendship and support to the school and our students throughout the years.”

This is the 14th year in which Madrona has formally recognized student research with commercial potential emerging from the Allen School.

Read more in the Madrona press release here, and check out GeekWire’s coverage of Althoff’s keynote here and the poster session here. See a complete list of past Madrona Prize winners here, and learn more about the Allen School’s Industry Affiliates program here.

Thanks to Madrona and to all of our industry partners, alumni and friends who showed up yesterday in support of our students, and congratulations to the winners — see you next year!

Read more →

 As part of the UW EXP Study led by professor Jennifer Mankoff in the Allen School’s Make4all Lab, a team of researchers set out to measure how specific incidents of discrimination can affect people’s behavior in the short-term by analyzing the experiences of college students at the University of Washington with the help of mobile and wearable devices.

In a paper presented last week at the Association for Computing Machinery’s 22nd Conference on Computer-Supported Cooperative Work and Social Computing (CSCW 2019) in Austin, Texas, Mankoff and her collaborators analyzed the prevalence of discriminatory events and their association  with behaviors among UW freshmen over two academic quarters in 2018. Working with 209 student volunteers — 176 of whom completed the study — the team sought to quantify each separate experience of discrimination to learn more about the immediate reactions. The study used passive sensing data collected via Fitbit Flex 2 wearable devices that tracked participants’ daily activities like sleep and movement and AWARE, an app installed on their smartphones that tracked their location, when and how often their phone screens were unlocked, and the length of their calls. The sensing data was correlated with the results of periodic surveys in which the students self-reported instances of unfair treatment. The researchers aimed to engage female students in STEM-related fields, where gender discrimination is an ongoing problem, as well as minorities and first-generation students in engineering. 

By the end of the study, participants had reported experiencing a total of 448 discriminatory events. The researchers found that, on average, those that reported being discriminated against were more physically active, interacted with their phones more, and slept less on the day of the event. 

“It’s so hard to summarize the impact of something like this in a few statistics,” Mankoff said in a UW News release. “Some people move more, sleep more or talk on the phone more, while some people do less. Maybe one student is reacting by playing games all day and another student put down their phone and went to hang out with a friend. It’s giving us a lot of questions to follow up on.”

Mankoff said the survey didn’t capture all discrimination events, but was instead a snapshot of some of the students’ experiences. More than half of them reported undergoing at least one discrimination event; many of those experienced about five events during the six month study period. In addition to the data on sleep, physical activity, and communication, the team found that discrimination events were associated with increased depression and loneliness — though the impact appeared to be less among those who could count on better social support.

“We looked at objective measures of behavior to try to really understand how this experience changed students’ daily life,” explained Allen School Ph.D. student Yasaman Sefidgar, lead author of the paper. “The ultimate goal is to use this information to develop changes that we can make both in terms of the educational structure and individual support systems for students to help them succeed both during and after their time in college.”

The researchers hope that, by possessing a better understanding of the consequences of discrimination, this work will lead to more supportive policies focused on prevention, intervention, and student retention in higher education.

“These students are not just giving us data, which sounds like some abstract, unemotional term,” Mankoff said. “They are sharing deeply personal information with us. It’s very important to me that we honor that gift by finding ways to help that don’t place the responsibility to deal with discrimination all on the individual.”

In addition to Mankoff and Sefidgar, co-authors on the paper include Allen School Professor Tim Althoff; UW Information School Dean Anind Dey; Eve Riskin, associate dean of diversity and access for the UW College of Engineering; Paula Nurius, professor of the UW School of Social Work; Anne Browning, founding director of the UW Resilience Lab; Kevin Kuehn, a clinical psychology doctoral student at the UW; and University of Michigan doctoral student Woo Suk Seo.

To learn more, read the full publication, “Passively-sensed behavioral correlates of discrimination events in college students,” the related UW News release, and coverage by Inside Higher Ed and The College Post. 

The University of Washington values and honors diverse experiences and perspectives, strives to create welcoming and respectful environments and promotes access and opportunity. Students, faculty and staff that encounter or suspect incidents of bias are encouraged to report it. 

Read more →

For students who are among the first in their families to attend college, the experience of navigating a four-year degree can be daunting. From decoding the campus lingo to overcoming imposter syndrome, the more than 250 undergraduates currently pursuing their bachelor’s degree as first-generation students in the Allen School are breaking down barriers and carving their own paths. We caught up with a few members of our community who are either finding their footing as first-generation students, or have been there, done that, and are happy to share what they learned in honor of the National First-Generation College Celebration taking place today at the University of Washington and across the nation.

Meet undergraduates Andres Eligio and Aaron Pham, graduate student Alyssa La Fleur, and academic adviser Chelsea Navarro — each with an inspirational story to tell about where they have been and where they are going as first-generation college students.

Andres Eligio

Andres Eligio is a freshman from Des Moines, Washington whose parents immigrated from Mexico in 1996 to provide a better life for their children. He is the first in his family to pursue an education after high school. Eligio credits the College Access Now (CAN) program as an important factor in his pursuit of a college degree. CAN helped him navigate and learn about colleges and the application process, while robotics, mathematics, and computer science classes in middle and high school solidified his interest in computer science.

Allen School: What does being a first-generation student mean to you?

Andres Eligio: To me it means facing the challenge of pursuing an education without having your parents’ support. It means working towards a goal which you can’t truly see. It can be scary to try and break away from what your parents have done. But I hope to fulfill all the work they did by coming to this country and take advantage of all the opportunities I am given. 

Allen School: What is your favorite part about being an Allen School student?

AE: My favorite part by far of the Allen School is how hard they work to make sure you feel like a part of the community. The faculty are very friendly and provide countless opportunities to make connections and learn more about being a UW student. I almost didn’t go to college. It took me a very long time to decide whether to continue my education after high school or work for my father’s landscaping company. I decided to try. I didn’t think I’d feel like I belonged, but I took part in the CSE Startup, a program for direct admit students that help them get used to life and classes at the UW. Before taking this course, I didn’t feel like I belonged. Now, having completed the program, I feel as if I am a part of the UW and more importantly, meant to be here in the Allen School. I feel confident about the rest of the school year and my success academically.

Allen School: What advice do you have for future first-gen students?

AE: I would tell them that I know it can be scary trying to pursue an education after high school. You can’t necessarily look to your parents for help, and they can only try to understand the struggle of finding the right college and choosing which field to study. Even though it seems hard and confusing, if you truly want to go to college you can do it. Look into your school or programs for support. Talk to counselors and friends. There will always be people to support you. It isn’t easy to leave your parents. Personally, it was really hard for me to leave. I worked hard to support my family, both in my dad’s business and in taking care of my brother. Many of you are in a similar situation, be it helping your parents or taking care of siblings. Your parents have worked really hard to provide you opportunities they didn’t have, and you should take full advantage of it.

Aaron Pham

Aaron Pham, a junior who transferred to the University of Washington in the spring, moved to the United States with his family in February of 2016. Born in Vietnam, he is the oldest child in his family and the first to go to college. His father worked for the U.S. Embassy in Vietnam; when the U.S. government gave him the opportunity to come to the States, the whole family moved to Washington, where Pham began his college career the following year at South Seattle College.

Allen School: What does being a first-generation student mean to you?

Aaron Pham: While it is such an honor, this also comes with challenges and responsibility. I will be a good example to my nephew and my younger cousins. It is a great motivation for me to keep learning, improving and pushing myself out of my own limits to become a successful student.

Allen School: What is your favorite part about being an Allen School student?

AP: My favorite part is the opportunities I have to connect with other friends and professors who also have a great passion for computer science. Studying and working in this environment not only improves my technical coding skills, but also guides me to become a person who wants to make impacts and contribute to the community and society by applying my knowledge and my passion for computer science. Being in the Allen School allows me to reach my full potential.

Allen School: What advice do you have for future first-gen students?

AP: Get involved at school, find your community, connect with a counselor, attend orientation activities, know where to get help on campus, and embrace who you are and don’t compare yourself to others. Everyone has their own weaknesses and strengths — so do you. No one is perfect. We are all here to learn and push ourselves out of our own limits. 

Alyssa La Fleur

Alyssa La Fleur, from Monroe, Washington, is a student in the Allen School’s full-time Ph.D. program. She fell in love with computational biology as an undergraduate and is now developing the skills to build a successful research career. La Fleur was homeschooled and attended a co-op before enrolling in Cascadia Community College through the Running Start program in high school. She graduated in the spring from Whitworth University with a triple major in math, bioinformatics and biochemistry. 

Allen School: What does being a first-generation student mean to you?

Alyssa La Fleur: It means that I will have greater job opportunities and financial security than my parents.

Allen School: What is your favorite part about being an Allen School student?

AL: So far, my favorite thing has been the friendly community and the diverse fields of study represented in it.

Allen School: What advice do you have for future first-gen students?

AL: Don’t be afraid to ask questions, even if you think they might be stupid. It’s also fine if you don’t know what questions you should be asking in the first place, as you are in a new environment and probably won’t realize what the gaps are in your knowledge right away. Also, if someone ever makes you feel uncomfortable when asking for help, there are plenty of other campus resources to use instead. I particularly recommend asking senior students in your major for advice.

Chelsea Navarro

Chelsea Navarro is an academic adviser at the Allen School focused on serving undergraduate students, including students transferring to UW from two-year colleges. As a first-generation student from San Diego, California, she credits services such as the Educational Opportunity Program (EOP) and Federal TRIO Programs in helping her begin her college career at Palomar Community College before transferring to San Diego State University, where she received her bachelor’s degree in sociology. The dedicated student affairs professionals and advisers that worked with her along the way inspired her to pursue a career in higher education. Navarro subsequently earned a Master’s of Education in student affairs from the University of California, Los Angeles and is proud of being a first-generation student. 

Allen School: What does being a first-generation student mean to you?

Chelsea Navarro: Growing up, my biggest ambition was to graduate from high school since it’s an accomplishment that not many people in my family are able to fulfill. My parents met as teenagers and had me when they were teens themselves. My father is a high school graduate and my mom dropped out of school when she was in middle school. I am the eldest of two daughters. One saying that has guided my practice is “remember why you started,” as so much of what I do is rooted in my higher education experience. I got into this field to help others and to hopefully be part of the support network that makes a student successful, like many advisers and faculty were for me when I was a student.

Allen School: What is your favorite part about working at the Allen School?

CN:  Working at the Allen School is an opportunity for me to continue to give back as faculty and advisers did for me. Part of my role as an undergraduate adviser at the Allen School is to work with our transfer students, which I absolutely enjoy doing. In many ways, it feels like I’ve come full circle and working at the Allen School is an amazing opportunity for me to continue to help others. 

Allen School: What advice do you have for future first-gen students?

CN:  When I first started at community college, I was incredibly lost and had a difficult time understanding university policies and interpreting my degree requirements: What is a credit? What happens in a lab? What is an associate’s degree? Is it okay to meet with professors? I’ll never forget the first time I met with an adviser and brought a huge list of questions with me to my appointment. I learned so much in those 30 minutes. Given my experience, my advice would be to encourage first-year, first-generation college students to be up front with the questions they want to have answered. It’s okay to admit that you feel lost and that you need help. As a first-generation student, I often felt like I was the only one going through the overwhelming experience of being the first in my family to go into higher education. Plus, I was scared to talk about my problems because in my mind, an administrator would notice and tell me that my greatest fear was true — they would say that I didn’t belong in higher education. Imposter syndrome is a difficult reality for many first-year, first-generation college students, so I encourage any students going through it to talk about it so they can get the support they need. 

We are grateful for the many contributions our first-generation students, faculty and staff have made to the Allen School community! Learn more about the National First-Generation College Celebration here, and activities celebrating UW’s first-generation community here.

Check out our student profiles from last year’s celebration here.

Read more →

Anyone who believes the adage “Rome wasn’t built in a day” hasn’t met the members of the Allen School’s Graphics and Imaging Laboratory (GRAIL). Ten years ago, postdoc Sameer Agarwal, Ph.D. student Ian Simon, alumnus Noah Snavely, professor Steve Seitz, and affiliate professor Richard Szeliski of Microsoft Research demonstrated how to digitally reconstruct the Italian capital in 3D using the large cache of photos shared on the internet. Last week, the team was one of two recipients of the Helmholtz Prize recognizing papers from a decade ago that have had a significant impact on computer vision research at the International Conference on Computer Vision (ICCV 2019) held in Seoul, Korea.

At the time the paper was written, city-scale 3D reconstructions largely relied on data from structured sources such as satellite imagery from Google Earth or street-level imagery captured by a moving vehicle. These visual datasets are typically produced by cameras with consistent calibration at a regular sampling rate. They are also often accompanied by additional sensor data, such as GPS, which further simplifies the computation involved in reconstructing a location. By contrast, images from unstructured sources — those posted on Flickr and other photo sharing websites — tend to share none of those characteristics. A search for “Rome” on Flickr returned more than two million photos at the time the researchers embarked on their project, reflecting a variety of camera settings, angles, lighting conditions, and location information.

To overcome these challenges and tap into what they described as “extremely rich source of information about the world,” the researchers sought to leverage massive parallelism along with the massive redundancy found in large internet photo collections. The team employed a combination of parallel distributed matching and reconstruction algorithms to create a system that scales in line with both the size of the problem and the available computational resources. Using this approach and applying state-of-the-art techniques such as structure from motion (SfM), SIFT, vocabulary trees, bundle adjustments, and more, they reconstructed a 3D version of the Eternal City in less than a day from a trove of 150,000 images harvested from the internet — the first city-scale reconstruction produced from unstructured photo collections.

The Helmholtz Prize is awarded every other year by the IEEE Computer Society’s Technical Committee on Pattern Analysis and Machine Intelligence. The team originally presented its winning paper at ICCV 2009 in Kyoto, Japan. Since then, Agarwal and Simon (Ph.D., ‘11) have gone on to engineering positions at Google, while Snavely (Ph.D., ‘08) is a member of the computer science faculty at Cornell Tech and a researcher at Google Research in New York City. Seitz currently splits his time between the Allen School and Google, where he serves as the director of teleportation, while Szeliski is now a research scientist and founding director of the Computational Photography Group at Facebook Research.

To learn more, read the winning research paper here, and check out the project website here.

Congratulazioni to the entire team!

Read more →

The strength and enduring impact of the Allen School and University of Washington’s contributions in accessible technology were on full display at the 21st International ACM SIGACCESS Conference on Computers and Accessibility, known as ASSETS, last month in Pittsburgh. Current or former Allen School researchers had a hand in three award-winning papers recognized at the conference, with a mix of current students, faculty, and alumni all represented. 

The Best Student Paper Award was granted to the paper “Deep Learning for Automatically Detecting Sidewalk Accessibility Problems Using Streetscape Imagery.” The authors,  Allen School Ph.D. students Galen Weld and Esther Jang, undergrad Aileen Zeng, professors Kurtis Heimerl, in the Information and Communication Technology for Development Lab  and Jon Froehlich, founder of the Makeability Lab and University of Maryland Ph.D. student Anthony Li, wrote about the use of deep learning to automatically assess the accessibility of sidewalks found in online imagery.

In an effort to let the public know which sidewalks in any city are accessible to people with disabilities, researchers began using smartphone-based tools to capture sidewalk accessibility problems on a large scale. Users marked accessibility on a map using their smartphones. This solution was not without flaws‒not many regions are adopting the program, the areas they cover are small and the burden on users is too much: they must download an app, take a picture, annotate it and upload it. To improve upon those issues, researchers implemented machine learning and satellite imagery to decrease manual labor and costs. However, even this work has been impacted negatively by small training sets and less than stellar machine-learning. To improve the process even more, the Allen School team used residual neural networks modified to support image and non-image features and had bigger training sets based on a dataset of 300,000+ image-based sidewalk accessibility labels collected by Project Sidewalk in the Makeability Lab. The results described in the winning paper were more accurate than previous automated methods, and in some cases exceeded the accuracy of human-generated labels.

Members of the Allen School also contributed to the 2019 SIGACCESS ASSETS Paper Impact Award, granted to a paper at least 10 years old that has had a significant impact on computing and information technology addressing the needs of persons with disabilities. Allen School  alumnus Jeffrey Bigham (Ph.D., ‘09), Information School professor and Allen School adjunct professor Jacob Wobbrock, and Information School alumnus Shaun Kane (Ph.D. ‘11), earned the award for their paper, “Slide Rule: Making mobile touch screens accessible to blind people using multi-touch interaction techniques.”  

Before the iPhone popularized the touchscreen smartphone, users who were blind could rely on their sense of touch using a cell phone’s physical buttons. To enable these users to take advantage of the latest smartphone features, the team designed Slide Rule, an audio-based program that enables blind users to access touch screen applications through the use of gestures. Slide Rule was the first prototype to exhibit finger-driven screen-reading and a second-finger tap gesture, today called “split tap,” to trigger on-screen targets like links and buttons. Since the team published its paper, major manufacturers have incorporated these features into their commercial products.

Last but not least, recent Allen School graduate Danielle Bragg (Ph.D., ‘18), who worked with Allen School professor Richard Ladner as a student, was lead author of the paper that earned  this year’s ASSETS Best Paper Award. Bragg and her co-authors at Microsoft Research were recognized for their paper “Sign language recognition, generation and translation: An interdisciplinary perspective,” which presents the results of an interdisciplinary workshop where researchers discussed developing successful sign language recognition, generation and translation systems in fields such as computer vision, computer graphics, natural language processing, human-computer interaction and linguistics. 

Congratulations to all of the ASSETS 2019 award recipients! 

Read more →

Researchers from the Allen School’s UNSAT group took home one of two Best Paper Awards and a Distinguished Artifact Award at the Association for Computing Machinery’s 27th Symposium on Operating Systems Principles (SOSP 2019) in Ontario, Canada last week. The winning paper, “Scaling symbolic evaluation for automated verification of systems code with Serval,” introduces a new framework for building automated verifiers for systems software.

Serval was developed by Allen School Ph.D. student and lead author Luke Nelson; alumnus James Bornholt (Ph.D., ‘19), now a faculty member at the University of Texas at Austin; Allen School professors Emina Torlak and Xi Wang; Columbia University professor Ronghui Gu; and Andrew Baumann, a principal researcher at Microsoft Research. Together, the researchers created a framework that overcomes several obstacles to scaling automated verification, including the developer effort required to write verifiers, the difficulty of finding and fixing performance bottlenecks, and limitations on their applicability to existing systems.

Unlike previous push-button verification approaches, which support automation at the expense of generality by requiring the co-design of systems and verifiers, Serval provides an extensible infrastructure that enables developers to easily retarget verifiers to new systems, including those not originally designed for automated verification. To do this, it leverages Rosette, a solver-aided programming language for synthesis and verification, to “lift” an interpreter into a verifier — that is, transform a regular program to work on symbolic values.

Verifiers created using Serval inherit a number of vital optimizations from Rosette, including constraint caching, state merging, and partial evaluation. But Serval goes a step further by introducing new capabilities for identifying and repairing performance bottlenecks. Employing recent advances in symbolic profiling, which offers a systematic approach to discovering performance bottlenecks, the researchers built a catalog of common bottlenecks for automated verifiers that includes indirect branches, memory accesses, trap dispatching, and more. They then built into Serval a set of symbolic optimizations that eliminate such bottlenecks and improve performance by exploiting domain knowledge to produce solver-friendly constraints for a class of systems.

To demonstrate Serval’s utility, the team developed reusable, interoperable automated verifiers for four instruction sets — RISC-V,  X86-32, LLVM, and Berkeley Packet Filter (BPF) — and used them to uncover previously unknown bugs in existing, unverified systems such as Keystone and the Linux kernel. They also applied Serval to two systems, CertiKOS and Komodo, to demonstrate how previously verified systems can be retrofitted for automatic verification.

Read the full research paper here, and explore the Serval on the project website here.

Way to go, team!

Read more →

Fall is back and so is the Allen School’s Undergrad Spotlight! This month’s student feature is Bellevue, Washington native Manoj Sarathy. Even before his arrival as part of the school’s expanded Direct to Major admissions program, the freshman computer science major was using machine learning to help environmental conservationists track and organize wildlife data. He was recently featured in the Seattle Times and on King 5 News for his work supporting wolverine recovery in Washington.  

Allen School: Why did you want to study computer science, and what made you choose the Allen School?

Manoj Sarathy: Like most high school seniors, I had a lot of interests but my work on applying machine learning to the field of environmental conservation showed me that computer science can be useful in essentially any field. I decided to study at the Allen School because of the connections the school has with all the major companies that are implementing machine learning — and I wanted to stay close to home in the beautiful Pacific Northwest.

Allen School: What do you enjoy most about being an Allen School student?

MS: I find the resources available to the Allen School’s undergraduate students to be extremely valuable. For example, the career fairs that took place earlier this month were very useful. I learned more about companies hiring in the computer science field.

Allen School: What activities and interests do you have outside of your studies?

MS: I have attended a meeting for the Society for Economic Restoration and will be attending some of their work parties to restore the campus. I am also interested in finding out more about Students Expressing Environmental Dedication (SEED). I hope to continue playing squash, a racquet sport, during my free time. One of the opportunities I gave up by coming to UW was playing for a varsity squash team, but I hope I can be in some kind of squash club here. 

Allen School: Why did you become a member of the Conservation Northwest while you were in high school?

MS: I wish I could say it was purposeful, but it was honestly an accident. I learned about the organization while doing some online research regarding environmental conservation in the Pacific Northwest for an environmental science class I was taking in high school. I really liked the work they do, like building wildlife overpasses and underpasses across I-90 and reintroducing fishers, a species that belongs in the same family as wolverines and which were wiped out in the Pacific Northwest by hunters. I reached out to the organization to learn about volunteer opportunities, and one thing led to another. At one time, I even printed t-shirts at home to raise funds for them and through that effort, met with some international conservation organizations.

One of the projects I became involved in was their camera trap project. Teams would hike up to areas where wildlife may be located to set up camera traps to observe predators and prey in that area. Conservation organizations use camera traps, but then have to spend a lot of time and work to classify the images. Involvement in that project led me to the idea of using machine learning to speed up that effort. 

Allen School: Is that when you began to work with Woodland Park Zoo’s senior conservation scientist, Robert Long?

MS: While working on my camera trap model, I quickly realized that I needed actual camera trap images from different cameras and angles to make my machine learning model accurate. I started writing to researchers who use camera traps and he was one of the few to respond immediately and generously offered his images to me to train my model. Luckily, he was in Seattle, and invited me to meet him at the Woodland Park Zoo. I have been working with him ever since. 

Allen School: How did you use machine learning to classify all of the images?

MS: Any machine learning system learns from input data. The better and the more varied the input data, the more accurate the machine learning system can be. Initially, I naively tried to use images from Google to train my machine learning model. I tried to create a model that distinguishes between species. When I tested the model with actual camera trap images, I quickly learned that the system was nearly useless because most images on the internet show animals in nearly ideal conditions, like with the background out of focus. Next, I found an online database used by prior researchers called “Snapshot Serengeti,” which has thousands of images of animals from Africa. Again, I found the lack of variety in the animals and vegetation to not be very useful for the camera trap images American conservationists were collecting.

I started writing to researchers and only a couple responded. Fewer still offered to share their images with me. I also learned through my discussions with them, and based on my own experience with Conservation Northwest’s camera trap project, that just separating images containing animals or humans from other images containing only background foliage would be immensely useful because researchers spend countless time looking at each false positive to make sure they are not missing anything. Distinguishing between animals and humans would also be very helpful. So I started building a model that classifies images into three categories: false positive, human, and animal. This enables volunteers to be more productive and efficient by prioritizing images for analysis. 

Allen School: Did you know how to do all of this before you started on the project? 

MS: Before I started working on my project, I knew next to nothing about coding and machine learning. I read as much as I could about machine learning and Google’s TensorFlow. I also needed to learn some Python programming to get it to work. Over time and through lots of failures and crashes, I slowly built a decent model. I don’t claim to understand how TensorFlow or machine learning frameworks actually work, but I hope to learn more about these topics in the Allen School! 

Allen School: Do you want to remain working in conservation after you finish your CS degree? 

MS: I genuinely enjoy the natural environment we are fortunate to have here in the Pacific Northwest. So I will definitely stay involved in environmental conservation, but I haven’t yet decided in what way or how I can make the most impact. Ask me that question again when I’m a senior, I may have a better idea.

We’re so excited to have a dedicated conservationist like Manoj as a member of the Allen School community. We are confident his innovation will change the world!

Read more →

Genetic genealogy websites enable people to upload their results from consumer DNA testing services like Ancestry.com and 23andMe to explore their genetic makeup, familial relationships, and even discover new relatives they didn’t know they had. But how can you be sure that the person who emails you claiming to be your Uncle Phil really is a long-lost relation?

Based on what a team of Allen School researchers discovered when interacting with the largest third-party genetic genealogy service, you may want to approach plans for a reunion with caution. In their paper “Genotype Extraction and False Relative Attacks: Security Risks to Third-Party Genetic Genealogy Services Beyond Identity Inference,” they analyze how security vulnerabilities built into the GEDmatch website could allow someone to construct an imaginary relative or obtain sensitive information about people who have uploaded their personal genetic data. 

Through a series of highly-controlled experiments using information from the GEDmatch online database, Allen School alumnus and current postdoctoral researcher Peter Ney (Ph.D., ‘19) and professors Tadayoshi Kohno and Luis Ceze determined that it would be relatively straightforward for an adversary to exploit vulnerabilities in the site’s application programming interface (API) that compromise users’ privacy and expose them to potential fraud. The team demonstrated multiple ways in which they could extract highly personal, potentially sensitive genetic information about individuals on the site — and use existing familial relationships to create false new ones by uploading fake profiles that indicate a genetic match where none exists.

Part of GEDmatch’s attraction is its user-friendly graphical interface, which relies on bars and color-coding to visualize specific genetic markers and similarities between two profiles. For example, the “chromosome paintings” illustrate the differences between two profiles on each chromosome, accompanied by “segment coordinates” that indicate the precise genetic markers that the profiles share. These one-to-one comparisons, however, can be used to reveal more information than intended. It was this aspect of the service that the researchers were able to exploit in their attacks. To their surprise, they were not only able to determine the presence or absence of various genetic markers at certain segments of a hypothetical user’s profile, but to reconstruct 92% of the entire profile with 98% accuracy.

As a first step, Ney and his colleagues created a research account on GEDmatch, to which they uploaded artificial genetic profiles generated from data contained in anonymous profiles from multiple, publicly available datasets designated for research use. By assigning each of their profiles a privacy setting of “research,” the team ensured that their artificial profiles would not appear in public matching results. Once the profiles were uploaded, GEDmatch automatically assigned each one a unique ID, which enabled the team to perform comparisons between a specific profile and others in the database — in this case, a set of “extraction profiles” created for this purpose. The team then performed a series of experiments. For the total profile reconstruction, they uploaded and ran comparisons between 20 extraction profiles and five targets. Based on the GEDmatch visualizations alone, they were able to recover just over 60% of the target profiles’ data. Based on their knowledge of genetics, specifically the frequency with which possible DNA bases are found within the population at a specific position on the genome, they were able to determine another 30%. They then relied on a genetic technique known as imputation to fill in the rest. 

Once they had constructed nearly the whole of a target’s profile, the researchers used that information to create a false child for one of their targets. When they ran the comparison between the target profile and the false child profile through the system, GEDmatch confirmed that the two were a match for a parent-child relationship.

While it is true that an adversary would have to have the right combination of programming skills and knowledge of genetics and genealogy to pull it off, the process isn’t as difficult as it sounds — or, to a security expert, as it should be. To acquire a person’s entire profile, Ney and his colleagues performed the comparisons between extraction and target profiles manually. They estimate the process took 10 minutes to complete — a daunting prospect, perhaps, if an adversary wanted to compare a much greater number of targets. But if one were to write a script that automatically performs the comparisons? “That would take 10 seconds,” said Ney, who is the lead author of the paper.

Consumer-facing genetic testing and genetic genealogy are still relatively nascent industries, but they are gaining in popularity. And as the size of the database grows, so does the interest of law enforcement looking to crack criminal cases for which the trail has gone cold. In one high-profile example from last year, investigators arrested a suspect alleged to be the Golden State Killer, whose identity remained elusive for more than four decades before genetic genealogy yielded a breakthrough. Given the prospect of using genetic information for this and other purposes, the researchers’ findings yield important questions about how to ensure the security and integrity of genetic genealogy results, now and into the future.

“We’re only beginning to scratch the surface,” said Kohno, who co-directs the Allen School’s Security and Privacy Research Lab and previously helped expose potential security vulnerabilities in internet-connected motor vehicles, wireless medical implants, consumer robotics, mobile advertising, and more. “The responsible thing for us is to disclose our findings so that we can engage a community of scientists and policymakers in a discussion about how to mitigate this issue.”

Echoing Kohno’s concern, Ceze emphasizes that the issue is made all the more urgent by the sensitive nature of the data that people upload to a site like GEDmatch — with broad legal, medical, and psychological ramifications — in the midst of what he refers to as “the age of oversharing information.”

“Genetic information correlates to medical conditions and potentially other deeply personal traits,” noted Ceze, who co-directs the Molecular Information Systems Laboratory at the University of Washington and specializes in computer architecture research as a member of the Allen School’s Sampa and SAMPL groups. “As more genetic information goes digital, the risks increase.”

Unfortunately for those who are not prone to oversharing, the risks extend beyond the direct users of genetic genealogy services. According to Ney, GEDmatch contains the personal genetic information of a sufficient number and variety of people across the U.S. that, should someone gain illicit possession of the entire database, they could potentially link genetic information with identity for a large portion of the country. While Ney describes the decision to share one’s data on GEDmatch as a personal one, some decisions appear to be more personal — and wider reaching — than others. And once a person’s genetic data is compromised, he notes, it is compromised forever. 

So whether or not you’ve uploaded your genetic information to GEDmatch, you might want to ask Uncle Phil for an additional form of identification before rushing to make up the guest bed. 

“People think of genetic data as being personal — and it is. It’s literally part of their physical identity,” Ney said. “You can change your credit card number, but you can’t change your DNA.”

The team will present its findings at the Network and Distributed System Security Symposium (NDSS 2020) in San Diego, California in February.

To learn more, read the UW News release here and an FAQ on security and privacy issues associated with genetic genealogy services here. Also check out related coverage by MIT Technology Review, OneZero, ZDNet, GeekWire, McClatchy, and Newsweek.

Read more →

Allen School Ph.D student Manaswi Saha has won an Amazon Catalyst award to support her research on “Combining computational and visualization techniques to understand urban accessibility at scale.” The award, which comes with $10,000 of funding attached, will support Saha’s dissertation research working with Allen School professor Jon Froehlich in the Makeability Lab.

“More than 30 million people have some form of disability in the U.S. Of these, half report using mobility aids. In spite of the growing need for accessible sidewalks, many cities remain inaccessible even after 25 years of the Americans with Disabilities Act regulations being in place,” Saha said. “Several cities have faced multi-million-dollar lawsuits for inaccessible sidewalks. However, there are currently no tools that can visualize and quantify this issue at scale.”

Saha’s latest endeavor is an extension of her work on Project Sidewalk, a web-based crowdsourcing tool. It gamifies the collection of data on curb ramps, obstacles and other relevant sidewalk conditions by allowing volunteers to virtually walk through online streetview imagery. The first deployment of the project in 2016 was done in Washington, D.C., where 797 online users audited 2,941 miles of streets to report on accessibility issues in the city, with subsequent deployments in Seattle and Newberg, Oregon. Saha will apply her Catalyst award towards building an interactive web visualization tool that will answer questions about accessibility for stakeholders: people with mobility disabilities, caregivers, local government officials (e.g. transportation departments), policymakers, and accessibility advocates. It will help answer questions such as: Which are the most inaccessible areas in the city? Why is my neighborhood inaccessible? Where should we prioritize for allocating resources for these repairs? 

The goals, Saha said, are to fill the informational gap between citizens and the local government in their understanding of urban accessibility, increase transparency by visualizing the current state of accessibility, and creating advocacy efforts for bringing about change.

“As a start, we will be utilizing data collected in D.C. from Project Sidewalk and other available data sources such as from the Department of Transportation,” Saha said. “Eventually, this work would be expanded to other cities to offer them similar support.”

Saha published “Project Sidewalk: A Web-based Crowdsourcing Tool for Collecting Sidewalk Accessibility Data at Scale” that earned a Best Paper Award at the Association for Computing Machinery Conference on Human Factors in Computing Systems (CHI 2019) in May. The co-authors include Froehlich, research scientist Michael Saugstad and undergrad Aileen Zeng at UW, students Hanuma Teja Maddali, Steven Bower, Aditya Dash and Anthony Li at the University of Maryland, College Park, high school student Ryan Holland, from Montgomery Blair High School, undergrad student Sage Chen from the University of Michigan and professor Kotaro Hara from Singapore Management University.

The Amazon Catalyst program is a collaboration between the University of Washington’s CoMotion and Amazon that grants funds to faculty, staff and students to encourage innovation. The goal is to support those in the UW community working on solutions to solve real-world problems. So far, the program has helped to fund 50 UW projects. Saha has been working on a novel solution to a real-world problem in urban transportation, an area of research the Catalyst program was focused on this year.

Read the Amazon Catalyst press release here, and learn more information about Project Sidewalk here. Check out past coverage of Saha and the Project Sidewalk team’s work by UW News, KIRO7, Crosscut, and Seattle Met.

Congratulations, Manaswi!

Read more →