Allen School News

All smiles: some of the Allen School’s newly-minted Ph.D.s

Tonight marked a special milestone in the history of Computer Science & Engineering at the University of Washington, as we celebrated the first graduating class of the Paul G. Allen School and recognized the students, faculty, alumni, and friends who make the Allen School community great.

We were honored to have engineer and entrepreneur Rich Barton — founder of Expedia, co-founder and executive chairman of Zillow, and co-founder and non-executive chairman of Glassdoor — join us to help mark the occasion as our inaugural graduation speaker. Barton kicked off the celebration with some advice for the graduates in a commencement speech that drew upon Greek mythology, history, and the Wizard of Oz. Pointing to the behavioral phenomenon known as the Pygmalion Effect — after the figure in an ancient Greek myth — he urged the new graduates to set great expectations, because “Great expectations beget great results….Have a dream, gather or join a talented crew of fellow adventurers, and make it so.”

And how do they do that? By emulating the Scarecrow, the Cowardly Lion, and the Tin Man in the Wizard of Oz. What those three wished for, said Barton, “are what I wish for you as you begin your journey in pursuit of your big, audacious dreams.” First, like the Scarecrow, graduates would need brains — plural, he pointed out, because in computer-science speak, “networks are much smarter, more complex, and interesting than nodes.” A diverse team accomplishes more collectively than an individual or a homogeneous team can. They would also need courage, like the Lion, and they should call upon it to “take big swings,” Barton said. “Take a chance on pursuing a big, audacious dream now.” Last — but certainly not least — is heart. He suggested that while fear may be easier than hope when it comes to inspiring others, the latter is both happier and healthier. “As you become leaders yourselves, remember to risk showing and sharing your own heart.”

Rich Barton advises graduates, “You have brains. Take courage. Display heart.”

GeekWire has a complete transcript of Barton’s address, plus a video.

Barton’s words set a terrific tone for the evening as we honored outstanding scholars and alumni who epitomize the Allen School’s commitment to educational excellence, service to our community, and generating real-world impact through research and innovation.

Each year, we recognize members of the graduating class who demonstrate superior scholarship, leadership potential, and the ability to apply and create new knowledge in the field of computing with Outstanding Senior Awards. In past years, we have awarded one award each to a student in computer science and in computer engineering. This year — as a reflection of how much our program has grown and the quality of our students — we decided to expand the award to two students in each discipline: Pascale Wallace Patterson and Sarah Yu in computer science, and Nick Anderson and Pooja Sethi in computer engineering.

Sethi also received the Best Senior Thesis Award, which recognizes the very best in undergraduate research at the Allen School based on originality, quality, and impact. She earned the award for her thesis “Respeak: A Voice-based, Crowd-powered, and Accessible Speech Transcription System,” which she completed working with professor Richard Anderson and Ph.D. student Aditya Vashistha. Again, we opted to break with tradition and recognize two students instead of one: Kuikui Liu also received a Best Thesis Award for “The Method of Interlacing Polynomials,” which he completed working with professors Shayan Oveis Gharan of the Allen School and Rekha Thomas of UW Mathematics.

We also recognized Royden Lucky with our Undergraduate Service Award, which is given to a graduating senior who has taken an active role in the Allen School community during his/her time as a student—contributing time, expertise, and enthusiasm to our events and activities.

The UW student chapter of the Association for Computing Machinery (ACM) recognized lecturer Adam Blank with its Undergraduate Teaching Award, highlighting his enthusiasm, his responsiveness, and his constant drive to improve the educational experience through innovation. Students themselves also can be powerful teachers — especially those who serve as teaching assistants (TAs). Each year we recognize and thank a small group of particularly outstanding TAs with our Bob Bandes Memorial Award. Our 2017 winners were Francis Ge, Varun Vijay Mahadevan, Alexander Tsun, Pascale Wallace Patterson, and Michelle Yun, with honorable mention going to Justin Huang and Evan McCarty.

A.J. Bernheim Brush (Ph.D. ’02); Hakim Weatherspoon (B.S. ’99)

Our new graduates are joining a long line of leaders and innovators who are using their UW education to change the world. To highlight this fact and recognize our most accomplished alumni, each year we announce two winners of our Alumni Achievement Award. This year’s honorees were A.J. Bernheim Brush (Ph.D., ’02) and Hakim Weatherspoon (B.S., ’99).

Brush is a leader in computer-supported collaborative work. After spending 11 years leading a variety of influential projects in Microsoft Research, she is now part of the Cortana product group working on far-field speech interaction, natural language processing, and user interface design for the company’s in-home digital personal assistant.

Weatherspoon combined Computer Engineering with varsity football at UW, traveling to four bowl games as a Husky. He followed his UW education with a Ph.D. from the University of California, Berkeley and has spent the past nine years as a professor at Cornell University. He is internationally renowned for his work on web-scale distributed systems and has earned accolades for his commitment to promoting diversity in computing.

Read more about Brush and Weatherspoon in our newsletter article here. Check out the complete graduation program here. And see GeekWire’s video of the entire ceremony here.

Congratulations to all of our graduates, and thanks to everyone who made our first Allen School graduation such a special occasion!

Event photo credits: Matt Hagen

  Read more →

Vincent Lee (left) and Max Willsey

Allen School Ph.D. students Vincent Lee and Max Willsey have been awarded a 2017 Qualcomm Innovation Fellowship for their proposal “Program Synthesis for Domain Specific Reconfigurable Accelerators.” Lee and Willsey, who were recommended by professors Ras Bodik, Luis Ceze, and Alvin Cheung, are one of only eight teams to receive a fellowship out of 33 finalists drawn from 116 original proposals received by the company.

Lee and Willsey are working with Bodik, Ceze, and Cheung on a project that aims to leverage solver-aided techniques and program synthesis ideas to design and build domain specific reconfigurable hardware accelerators. The goal of their research, which spans computer architecture, compilers, and programming languages, is to achieve significant performance improvements while simultaneously reducing the energy costs associated with computation. The project is housed in the Allen School’s inter-disciplinary SAMPA group, which advances research across multiple layers of the system stack, including hardware, programming languages, and applications.

The Qualcomm Innovation Fellowship program is designed to recognize and support Ph.D. students and their forward-thinking research in a variety of technical areas that are driving computing innovation. This is the fifth year in a row that an Allen School team has earned one of these coveted awards and the second win for Lee, who received a fellowship in 2015 with Carlo del Mundo for their work to develop systems and architecture support for large-scale video search. Winning teams each receive a $100,000 fellowship and mentorship by Qualcomm engineers.

Congratulations, Vincent and Max! Read more →

The 2017 graduation ceremony of the Paul G. Allen School of Computer Science & Engineering will be livestreamed – 6:00 p.m. on Friday June 9, beginning with an address to the graduates by Zillow Group co-founder and Executive Chairman Rich Barton. Other highlights include presentation of the 2017 Alumni Achievement Awards to A.J. Brush (Ph.D. ’02) and Hakim Weatherspoon (B.S., ’99), various other awards, the procession of Bachelors and Masters graduates, and the hooding of Bachelors graduates.

Watch at https://www.youtube.com/uwcse!

Read more →

Allen School students and faculty at SIGMOD in Chicago

A delegation of Allen School researchers recently returned from the annual conference of the Association for Computing Machinery’s Special Interest Group on Management of Data (SIGMOD 2017). Several faculty and students from the Database and Programming Languages & Software Engineering groups were recognized for their work to advance the state of the art in data management.

Ph.D. student Shumo Chu earned the Best Demo Award for his presentation of the Cosette Automated SQL Prover, a project developed by Chu, his fellow Ph.D. student Chenglong Wang, undergraduate researcher Daniel Li, and professors Alvin Cheung and Dan Suciu. Cosette is the first automated prover for determining the equivalence of two SQL queries. The team has demonstrated its usefulness by proving a number of real-world SQL rewrite rules ​that have not been shown correct before, along with disproving well-known bugs ​in SQL query optimizers, ​all in a matter of seconds. The team ​will  release the tool in the next few weeks.

Shumo Chu demonstrating Cosette

Two other students received Honorable Mentions for their demos. Brandon Haynes earned accolades for demonstrating VisualCloud, a database management system for efficient storage and delivery of virtual reality content at scale that he developed with 5th year master’s student Artem Minyaylov and professors Cheung, Magdalena Balazinska, and Luis Ceze. Maaz Bin Safeer Ahmad was recognized for his demo of Casper, a tool he developed with Cheung that automatically rewrites sequential Java applications to leverage parallel data processing frameworks such as Spark and Hadoop using verified lifting.

In the ACM Student Research Competition, Ph.D. student Jennifer Ortiz was named First Runner-Up in the graduate student category for her poster and research talk on PerfEnforce, a dynamic scaling engine for analytics with performance guarantees. Ortiz developed PerfEnforce in collaboration with Balazinska, master’s alum Brendan Lee, Microsoft Technical Fellow Johannes Gehrke, and Senior Data Science Fellow Joseph Hellerstein of the UW eScience Institute.

A team of Allen School researchers also earned one of three Reproducibility Awards, which recognize papers from the previous year’s conference that are deemed to be most reproducible, verifiable, flexible, and portable. The winning paper, “SQLShare: Results from a Multi-Year SQL-as-a-Service Experiment,” presented the results of a four-year deployment of a database-as-a-service platform aimed at scientists with minimal database experience and was co-authored by Ph.D. students Shrainik Jain and Dominik Moritz, Ph.D. alum Daniel Halperin, iSchool professor and Allen School adjunct faculty member Bill Howe, and Allen School professor Ed Lazowska.

Professor Alvin Cheung (left) with ACM Student Research Competition honorees Utku Sriin, Jennifer Ortiz, and Lingjiao Chen

Last but certainly not least, Balazinska collected SIGMOD’s Test of Time Award for her 2005 paper “Fault Tolerance in the Borealis Distributes Stream Processing System,” which pioneered a new approach for increasing the fault tolerance of stream processing applications. The award, which Balazinska shares with co-authors Hari Balakrishnan, Samuel Madden, and Michael Stonebraker of MIT, signals the enduring influence her work has had on the field of data management in the dozen years since its initial publication. Read more in our previous blog post here.

View more photos of the Allen School in action at SIGMOD here.

Great work, everyone! Read more →

This evening, Amazon Professor of Machine Learning Emily Fox was recognized by the Seattle chapter of the Association for Women in Science (AWIS) with their 2017 Award for Scientific Achievement in STEM.

AWIS writes: “Emily is an expert in machine learning and a leading researcher in redefining the scope and nature of applied statistics. She is a leader in developing computationally realistic modeling tools for complex data sets. In addition to teaching and advising at the University of Washington, she co-created an online course about machine learning. She was recently recognized by President Obama with the Presidential Early Career Award for Scientists and Engineers. Emily fosters the development of other women in science through her visible success and being approachable and available as a mentor and counselor.”

Congratulations Emily! Read more →

Tanner Schmidt (left) speaks to attendees at ICRA

A team of researchers in the Allen School’s Robotics and State Estimation Lab earned the award for Best Robotic Vision Paper at the recent IEEE International Conference on Robotics and Automation (ICRA 2017). The winning paper, “Self-supervised Visual Descriptor Learning for Dense Correspondence,” represents a significant step forward in robot learning by providing a framework for enabling robots to understand their local environments without human intervention.

In order for robots to operate safely and effectively in dynamic environments, they must be able to recognize objects and scenes they have encountered on previous occasions and apply that knowledge to their current situation. The process by which robots develop this human-like perception of their environment is known as correspondence estimation, which typically requires image descriptors that have been engineered by hand or the expensive collection of vast quantities of training data to take advantage of deep learning techniques.

The team from the RSE-Lab — Allen School Ph.D. student Tanner Schmidt, former postdoc and current affiliate professor Richard Newcombe of Oculus, and professor Dieter Fox — came up with an alternative that automates the generation of training data and enable robots to learn the visual features of a scene in a self-supervised way. Leveraging dense mapping techniques such as KinectFusion and DynamicFusion, the researchers were able to generate correspondence labels from raw RGB-D video data. The researchers then used the resulting labels and a contrastive loss to train a fully convolutional network to produce dense visual descriptors from novel images that are consistent despite variations in pose, viewpoint, or lighting conditions. This work, which represents the state of the art in descriptor learning, will be useful for researchers tackling a number of important problems in robot vision, including tracking, mapping, and object recognition.

Another Allen School paper, “SE3-Nets: Learning Rigid Body Motion using Deep Neural Networks” by Ph.D. student Arunkumar Byravan and Fox, was a finalist for the same award. That paper describes SE3-Nets, which are deep neural networks for modeling and predicting the motion of objects subject to applied force. Byravan and Fox demonstrated that SE3-Nets are able to learn scene dynamics from limited real-world data, generalize across different scenes, and more consistently predict object motion compared to traditional flow networks.

As Fox points out, vision is becoming an increasingly important area of robotics — and our showing at ICRA demonstrates that we are at the forefront of this exciting line of research. Way to go, team! Read more →

Richard Ladner and some of his students who joined the dinner in his honor. Back row: Sangyun Hahn, Kevin Zatloukal, Albert Greenberg, Richard Ladner, Shiri Azenkot, Catherine Baker, Danielle Bragg, Jeff Bigham, Soma Chaudhuri. Middle row: Lauren Milne, Anna Cavender, Cynthia Bennett. Front row: Anne Condon, Shaun Kane, Jessica Tran.

Richard Ladner joined the Computer Science Group at the University of Washington in 1971 – 45 years ago this past fall.

Richard was first a leader in theoretical computer science and then a leader in accessibility. He has supervised or co-supervised 27 Ph.D. students with 4 more in the pipeline. He has also supervised the research of more than 100 undergraduate students, 23 of whom won Mary Gates Research Scholarships, and 2 who won CRA Outstanding Undergraduate Researcher Awards. He served as Chair of the ACM Special Interest Group in Algorithms and Computation Theory (SIGACT) from 2005-2009. He was a member of the Board of Trustees of Gallaudet University from 2007-2016. He is a recipient of the 2004 Presidential Award for Excellence in Science, Mathematics and Engineering Mentoring (PAESMEM) and the 2008 A. Nico Habermann Award. He is the winner of the 2014 SIGCHI Social Impact Award and 2016 SIGACCESS Award for Outstanding Contributions to Computing and Accessibility. He is an ACM Fellow and an IEEE Fellow. He was a Guggenheim Fellow in 1986 and Fulbright Scholar in 1993. At the University of Washington he received the Boeing Professorship in Computer Science & Engineering (2004-2012), the Undergraduate Research Mentor Award (2010), the Samuel E. Kelly Distinguished Faculty Lecture Award (2010), and the University of Washington Outstanding Service Award (2009).

On Friday we celebrated his career on the occasion of his retirement, first with a valedictory lecture, and then with a dinner in his honor.

Richard’s former students are establishing the Richard Ladner Endowed Graduate Fellowship in his honor: “to support students studying accessible technology in the Allen School, the Information School or the College of Engineering, or students with disabilities in those schools and colleges whose research focus is a field within computer science.” If you would like to support the Ladner Fellowship, you can find information here.

A profile of Richard, excerpted from the Summer 2017 issue of CSE’s newsletter Most Significant Bits, is here.

The slides from Richard’s valedictory lecture, “My 45+ Years at UW,” are available in pptx and pdf. The talk announcement (with a link to the video) is here. Read more →

“Shyam Gollakota envisions a future in which his research and technological breakthroughs will transform computing as we know it. It’s a powerful prediction, and it’s based on the consumption of less power.

“Gollakota, a co-founder of Jeeva Wireless and an assistant professor in Computer Science & Engineering at the University of Washington, is a leader in the field of wireless networking. He was named one of MIT Technology Review’s innovators under 35 in 2014; part of the Popular Science “Brilliant 10” in 2016; and was among the Forbes “30 under 30” in 2015 and 2017.

“While those all sound pretty fancy, we’re here to name him GeekWire’s latest Geek of the Week.”

Shyam joins previous Allen School “Geeks of the Week” Irene Zhang, Karan Goel, Carlos Guestrin, Julie Kientz, Melissa Winstanley, Lauren Bricker, Yaw Anokwa, Wendy Chisholm, and Marty Stepp, as well as 2012 “Geek of the Year” Oren Etzioni and 2017 “Geek of the Year” Ed Lazowska.

We got geek! Read more here. Read more →

Our latest installment of the Allen School’s Undergrad Spotlight features Colin Summers, a senior majoring in computer engineering with significant experience in chemical and mechanical engineering. The Seattle native, who is pursuing a minor in aeronautics and astronautics, will spend the summer as an intern at the National Aeronautics and Space Administration’s Jet Propulsion Laboratory (NASA JPL) in California.

Before taking off on his new adventure, Summers took time out from his busy schedule to share with us what inspires him as a student, what he learned while working on the award-winning UW Hyperloop project, and what’s next.

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

Colin Summers: The excitement of the students. Of every academic program that I have been a part of — and there have been quite a few — I have not seen students more self-motivated and passionate than CSE students. Everywhere you look, students are chasing their interests whether it be robotics, video games, travel, theory, or any one of the wild things people pursue in this school.

Allen School: Who has inspired you the most as a student?

CS: There’s almost too many to list! Some recent inspirations might include fellow students Johan Michalove for sharing a giddy excitement over robotics, Meredith Lampe for helping to convince me to give CSE 142 a shot, and John Kaltenbach for continually stoking my love of travel; professor Dieter Fox, for sharing a glimpse of the future of robotics; Adam Blank, for demonstrating how to be yourself in the pursuit of what you love; Jonathan Lee, for getting me excited about computational theory and mathematics all over again…As I said, too many to list!

Allen School: What hobbies and interests do you have outside of school?

CS: I’m an avid skier and climber and have recently started to merge the two in the pursuit of mountaineering. I love travel and have spent a bit over a year living in South America, plus a bit of time in Indonesia.

Allen School: What is one of your favorite student memories so far?

CS: Flying to Texas last year with the UW Hyperloop team for the Design and Build competition was one of the most incredible experiences that I’ve had in college. There was about 15-20 of us that went down to pitch our design to a panel of over 100 judges from industry and academia, in the hopes of securing a spot in the final round of the competition. Of over 1,200 teams that applied, only 120 were selected to present — and fewer than 30 teams made it out with approval to build out their designs and race.

The competition is now over, and UW Hyperloop placed 4th in the US and 6th in the world. Hyperloop presented the opportunity to work on a radical new concept that had never been done before, alongside some of the most excellent engineers I’ve ever worked with. It challenged me to think outside the box and to recognize the importance of team dynamics and vision in leading an engineering project.

Allen School: How do you follow up an amazing experience like that — what’s next for you?

CS: Recently, Johan Michalove and I began a robotics project, Dub Car, as part of a new undergraduate program titled Cyber-Physical Systems Group (CPSG). What first started as a side project has since matured into a full-blown honors research project under professor Dieter Fox. Our task is to design, build, and program a 1/10th scale autonomous race car from commercial parts that can be used as a future platform for research in control theory, computer vision, systems development, and autonomy.

Through sponsorship, donations, and our own pockets we’ve managed to scrape together all of the parts needed for our platform and are in the final stages of assembly. A lot of work remains (especially if I keep blowing up the motor controller!), but we’re in it for the long haul and have loved every minute of it thus far. Luckily, we somehow both managed to land internships at NASA JPL for this summer, and we are excited to keep the progress going year-round!

We’re happy to have a star like Colin as part of the Allen School community. Good luck in your internship, Colin! Read more →

Researchers Peter Ney (left) and Ian Smith

A team of researchers in the Allen School’s Security and Privacy Research Lab have developed a new system, SeaGlass, which could bring more transparency and accountability to cell-phone surveillance. SeaGlass is capable of detecting anomalies in the cellular network that may indicate the presence of surveillance devices called IMSI-catchers (also known as cell-site simulators or Stingrays), which track individuals through their International Mobile Subscriber Identity by posing as a legitimate cell tower.

WIRED magazine recently talked to the researchers about SeaGlass and their hope that the findings contribute to the public discourse. From the article:

“Law enforcement’s use of the surveillance devices known as stingrays, fake cell towers that can intercept communications and track phones, remains as murky as it is controversial, hidden in non-disclosure agreements and cloak-and-dagger secrecy. But a group of Seattle researchers has found a new method to track those trackers: by recruiting ridesharing vehicles as surveillance devices of their own.

“For two months last year, researchers at the University of Washington paid drivers of an unidentified ridesharing service to keep custom-made sensors in the trunks of their cars, converting those vehicles into mobile cellular data collectors.”

Data collected from a single cell tower in Seattle over two months

The article explains how the information gathered by those mobile data collectors was analyzed to identify irregularities in the network compared to normal cell tower behavior, which could indicate that a surveillance device is active in the area. By using off-the-shelf parts and deploying the system on fleet vehicles that cover a lot of ground during their normal course of business, the researchers demonstrated SeaGlass to be an inexpensive and unobtrusive way to map the cellular landscape.

During the pilot, the system detected three anomalies in the greater Seattle area that piqued the team’s interest. Although the results do not offer conclusive proof of an IMSI-catcher, they do provide a good starting point for further exploration — and for an overdue conversation about surveillance practices.

“Up until now the use of IMSI-catchers around the world has been shrouded in mystery, and this lack of concrete information is a barrier to informed public discussion,” Ph.D. student Peter Ney said in a UW News release. “Having additional, independent and credible sources of information on cell-site simulators is critical to understanding how — and how responsibly — they are being used.”

The team, which also includes Allen School research scientist Ian Smith, Ph.D. student Gabriel Cadamuro, and professor Tadayoshi Kohno, hopes that its work will contribute to a more robust public debate over cell-phone surveillance. They describe in detail the SeaGlass technology and results of the pilot in a paper that will be published this month in the Proceedings on Privacy Enhancing Technologies.

“SeaGlass is a promising technology that — with wider deployment — can be used to help empower citizens and communities to monitor this type of surveillance,” said Smith.  “This issue is bigger than one team of researchers. We’re eager to push this out into the community and find partners who can crowdsource more data collection and begin to connect the dots in meaningful ways.”

Read the WIRED story here, the UW News release here, and additional coverage by TechCrunch, Gizmodo, and Engadget. Learn more at the SeaGlass website here. Read more →