Skip to main content

Allen School’s Amy Zhang and Franziska Roesner win NSF Convergence Accelerator for their work to limit the spread of misinformation online

Amy Zhang (left) and Franziska Roesner

The National Science Foundation (NSF) has selected Allen School professors Amy Zhang, who directs the Social Futures Lab, and Franziska Roesner, who co-directs the  Security and Privacy Research Lab, to receive Convergence Accelerator funding for their work with collaborators at the University of Washington and the grassroots journalism organization Hacks/Hackers on tools to detect and help stop misinformation online. The NSF’s Convergence Accelerator program is unique in that its structure offers researchers the opportunity to accelerate their work over the course of a year to find tangible solutions. The curriculum is designed to strengthen each team’s convergence approach and further develop their solution to move on to a second phase with the potential for additional funding.

In their proposal, “Analysis and Response for Trust Tool (ARTT): Expert-Informed resources for Individuals and Online Communities to Address Vaccine Hesitancy and Misinformation,” Zhang, Roesner, Human Centered Design & Engineering professor and Allen School adjunct professor Kate Starbird, Information School professor and director of the Center for an Informed Public Jevin West, and internet and Hacks/Hackers researcher at large Connie Moon Sehat, who serves as principal investigator on the project, aim to develop a software tool — ARTT — that helps people identify and prevent misinformation. This currently happens on a smaller scale by individuals and community moderators with few resources or expert guidance on combating false information. The team, made up of experts in fields such as computer science, social science, media literacy, conflict resolution and psychology, will develop a software program that helps moderators analyze information online and present practical information that builds trust.  

“In our previous research, we learned that rather than platform interventions like ‘fake news’ labels, people often learn that something they see or post on social media is false or untrustworthy from comment threads or other community members,” said Roesner, who serves as co-principal investigator on the ARTT project alongside Zhang. “With the ARTT research, we are hoping to support these kinds of interactions in productive and respectful ways.”

While ARTT will help prevent the spread of any misinformation, the team’s focus right now is on combating false information on vaccines — vaccine hesitancy has been identified by the World Health Organization as one of the top 10 threats to global health.

In addition to her participation in the ARTT enterprise, Zhang has another Convergence Accelerator project focused on creating a “golden set” of guidelines to help prevent the spread of false information. That proposal, “Misinformation Judgments with Public Legitimacy,” aims to use public juries to render judgments on socially contested issues. The jurors will continue to build these choices to create a “golden set” that social media platforms can use to evaluate information posted on social media. Besides Zhang, the project team includes the University of Michigan’s Paul Resnick, associate dean for research and faculty affairs and professor at the School of Information, and David Jurgens, professor at the Information School and in the Department of Electrical Engineering & Computer Science, and the Massachusetts Institute of Technology’s David Rand, professor of management science and brain and cognitive sciences and Adam Berinsky, professor of political science.

Online platforms have been increasingly called on to reduce the spread of false information. There is little agreement on what process should be used to do so, and many social media sites are not fully transparent about their policies and procedures when it comes to combating misinformation. Zhang’s group will develop a forecasting service to be used as external auditing for platforms to reduce false claims online. The “golden sets” created from the jury’s work will serve as training data to improve the forecasting service over time. Platforms that use this service will also be more transparent about their judgments regarding false information posted on their platform. 

“The goal of this project is to determine a process for collecting judgments on content moderation cases related to misinformation that has broad public legitimacy,” Zhang said. “Once we’ve established such a process, we aim to implement it and gather judgments for a large set of cases. These judgments can be used to train automated approaches that can be used to audit the performance of platforms.”

Participation in the Convergence Accelerator program includes a $749,000 award for each team to develop their work. Learn more about the latest round awards here and read about all of the UW teams that earned a Convergence Accelerator award here

Read more →

Professor Franziska Roesner earns Consumer Reports Digital Lab Fellowship to support research into problematic content in online ads

Franziska Roesner smiling and leaning against a wood and metal railing
Credit: Dennis Wise/University of Washington

As anyone who has visited a website knows, online ads are taking up an increasing amount of page real estate. Depending on the ad, the content might veer from mildly annoying to downright dangerous; sometimes, it can be difficult to distinguish between ads that are deceptive or manipulative by design and legitimate content on a site. Now, Allen School professor Franziska Roesner (Ph.D., ‘14), co-director of the University of Washington’s Security and Privacy Research Lab, wants to shed light on problematic content in the online advertising ecosystem to support public-interest transparency and research.

Consumer Reports selected Roesner as a 2021-2022 Digital Lab Fellow to advance her efforts to create a public-interest online ads archive to document and investigate problematic ads and their impacts on users. With this infrastructure in place, Roesner hopes to support her team and others in developing new user-facing tools to combat the spread of misleading and potentially harmful ad content online. She is one of three public interest technology researchers to be named in the latest cohort of Digital Lab Fellows focused on developing practical solutions for addressing emerging consumer harms in the digital realm. 

This is not a new area of inquiry for Roesner, who has previously investigated online advertising from the perspective of user privacy such as the use of third-party trackers to collect information from users across multiple websites. Lately, she has expanded her focus to examining the actual content of those ads. Last year, amidst the lead-up to the U.S. presidential election and the pandemic’s growing human and economic toll — and against the backdrop of simmering arguments over the origins of SARS-CoV-2, lockdowns and mask mandates, and potential medical interventions — Roesner and a team of researchers unveiled the findings of a study examining the quality, or lack thereof, of ads that appear on news and media sites. They found that problematic online ads take many forms, and that they appeared equally on both trusted mainstream news sites and low quality sites devoted to peddling misinformation. In follow-up work, Roesner and her collaborators further studied people’s — not just researchers’ — perceptions of problematic ad content, and in forthcoming work, problematic political ads surrounding the 2020 U.S. elections.

“Right now, the web is the wild west of advertising. There is a lot of content that is misleading and potentially harmful, and it can be really difficult for users to tell the difference,” explained Roesner. “For example, ads may take the form of product ‘advertorials,’ in which their similarity to actual news articles lends them an appearance of legitimacy and objectivity. Or they might rely on manipulative or click-baity headlines that contain or imply disinformation. Sometimes, they are disguised as political opinion polls with provocative statements that, when you click on them, ask for your email address and sign you up for a mailing list that delivers you even more manipulative content.”

Roesner is keen to build on her previous work to improve our understanding of how these tactics enable problematic ads to proliferate — and the human toll that they generate in terms of the time and attention wasted and the emotional impact of consuming misinformation. Building out the team’s existing ad collection infrastructure, the ad archive will provide a structured, longitudinal, and (crucially) public look into the ads that people see on the web. These insights will support additional research from Roesner’s team as well as other researchers investigating how misinformation spreads online. Roesner and her collaborators ultimately aim to help “draw the line” between legitimate online advertising content and practices, and problematic content that is harmful to users, content creators, websites, and ad platforms.

But Roesner doesn’t think we should wait for the regulatory framework to catch up. One of her priorities is to protect users from problematic ads, such as by developing tools that automatically block certain ads or empower users to recognize and flag them. While acknowledging that online advertising is here to stay — it funds the economic model of the web, after all — Roesner believes that there is a better balance to be struck between revenue and the quality of content that people consume on a daily basis as they point and click.

“Even the most respected websites may be inadvertently hosting and assisting the spread of bogus content — which, as things stand, puts the onus on users to assess the veracity of what they are seeing,” said Roesner. “My hope is that this collaboration with Consumer Reports will support efforts to analyze ad content and its impact on users — and generate regulatory and technical solutions that will lead to more positive digital experiences for everyone.”

Consumer Reports created the Digital Lab Fellowship program with support from the Alfred P. Sloan Foundation and welcomed its first cohort last year. 

“People should feel safe with the products and services that fill our lives and homes. That depends on dedicated public interest technologists keeping up with the pace of innovation to effectively monitor the digital marketplace,” Ben Moskowitz, director of the Digital Lab at Consumer Reports, said in a press release. “We are proud to support and work alongside these three Fellows, whose work will increase fairness and trust in the products and services we use everyday.”

Read the Consumer Reports announcement here, and learn more about the Digital Lab Fellowship program here.

Congratulations, Franzi!

Read more →

“There’s so much beauty in these tiny things”: Allen School’s Shyam Gollakota named 2021 Moore Inventor Fellow for advancing big ideas on a miniature scale

Portrait of Shyam Gollakota
Credit: Tara Gimmer

Allen School professor Shyam Gollakota has received a 2021 Moore Inventor Fellowship in recognition of his work at the nexus of low-power wireless communication, biology and living organisms. Gollakota, who directs the Allen School’s Networks & Mobile Systems Lab, is the first University of Washington faculty member to receive this prestigious award that nurtures the next generation of scientist-inventors. The Gordon and Betty Moore Foundation established the fellowship program aimed at supporting “50 inventors to shape the next 50 years” in 2016 to mark the 50th anniversary of Moore’s Law positing the exponential growth in computer processing power. 

Power figures prominently in Gollakota’s work — but in his case, the focus has been on finding ways to wirelessly fuel computation in order to cut the cord and lighten the load. Much like the results of the eponymous law articulated by Gordon Moore, Gollakota’s research has led to expanded computational capabilities accompanied by a shrinking form factor.

His first foray into wireless computing resulted in a breakthrough known as ambient backscatter. Together with his UW colleague Joshua Smith, who holds a joint appointment in the Allen School and the Department of Electrical & Computer Engineering, Gollakota developed a battery-free system that used television, WiFi and other wireless signals as both a power source and a mode of communication. In a series of subsequent projects, Gollakota and his collaborators expanded these capabilities to cover greater distances and bestow the capability to perform wireless computation on a greater variety of objects.

After looking skyward to enable devices to pull power out of thin air, Gollakota cast his eyes in the opposite direction as he contemplated how to make the most of these new capabilities.

“Outside there’s a whole world on every square foot, with living beings that you don’t even think about. We just walk over it,” Gollakota said in a UW News story. “But there’s so much happening — feats of engineering. There’s so much beauty in these tiny things.”

Gollakota and his colleagues drew inspiration from those “tiny things” to engineer a new line of research he has dubbed the Internet of Biological and Bio-Inspired Things. The concept began to take off with the development of a lightweight wireless sensor backpack small enough to be carried by bumblebees. The onboard sensors gather data about the surrounding environment as the bees go about their daily business; upon their return to the hive each evening, the data they logged is uploaded using backscatter while the tiny battery is wirelessly recharged for the next day’s flight. Gollakota and his collaborators followed up that buzz-worthy project with a wireless sensing package that could be safely air-dropped from great heights by live moths or drones into remote or impassable areas, followed by a miniature remote-control camera that can ride on the back of a beetle. Looking to the future, Gollakota is keen to explore ways to more deeply integrate biology and technology to achieve his vision.

Bumblebee wearing tiny sensor on its back collecting nectar from a flower
Credit: Mark Stone/University of Washington

“Simply put, Shyam is amazing — he is easily the most creative person I have ever met,” his Allen School colleague Thomas Anderson observed earlier this year. “He repeatedly invents and builds prototypes that, before you see them demonstrated, you would have thought impossible.”

While Gollakota’s notion of an Internet of Biological and Bio-Inspired Things may at first seem to belong in the realm of science fiction, it has many practical applications, from wildlife conservation, to smart agriculture, to large-scale environmental monitoring. In parallel with this work, Gollakota has also collaborated with colleagues and clinicians on a series of mobile sensing projects to support contactless disease detection and health monitoring using smartphones and smart speakers.

Gollakota is one of five innovators to be named in the 2021 cohort of Moore Inventor Fellows. He and his fellow honorees were selected from nearly 200 nominations received by the Foundation and will each receive $825,000 to further their inventions. Gollakota, who holds the Torode Family Career Development Professorship in the Allen School, previously earned the Association for Computing Machinery’s ACM Grace Murray Hopper Award in recognition of his early-career technical contributions, MIT Technology Review’s TR35 Award recognizing the world’s top innovators under the age of 35, and a Sloan Research Fellowship — among many other honors since his arrival at the UW in 2012.

Read the Moore Foundation announcement here, Gollakota’s Moore Inventor Fellow profile here, and a related UW News story here.

Congratulations, Shyam!

Read more →

Not just phoning it in: Shwetak Patel honored by Georgia Tech and Business Insider for contributions to low-power sensing and mobile health innovation

Shwetak Patel with arm extended toward camera and color calibration card resting on forearm while a pair of hands extends from off camera holding a cell phone

University of Washington professor Shwetak Patel has earned a place in the Georgia Tech College of Computing’s Hall of Fame and a spot on Business Insider’s recent list of “30 leaders under 40” who are changing health care for his innovative work combining low-power sensing, signal processing and machine learning for applications ranging from non-invasive disease screening to monitoring appliance-level energy consumption. Patel, who holds the Washington Research Foundation Entrepreneurship Endowed Professorship in the Allen School and the UW Department of Electrical & Computer Engineering, is also a serial entrepreneur and director of health technologies at Google Health and Fitbit Research.

“Ten years ago, computing’s role in health care was basically billing, data collection and databases,” Patel observed to Business Insider. “Now computing is playing a critical role in the actual discovery of new interventions in health outcomes.” 

Patel himself has been largely responsible for transforming computing’s contribution from staid billing software to pocket-sized personal health monitor. Evidence of that work is strewn around the Allen School’s Ubiquitous Computing Lab on the UW’s Seattle campus, where stacks of mobile phones and coils of charger cable jostle for space with a variety of 3D-printed accessories, camera color correction cards, the odd blood pressure monitor, and even a life-sized plastic baby doll. (That last one is used for demonstrating an app for detecting infant jaundice.)

Patel’s drive to “democratize diagnostics,” as he once described it, stemmed from his realization that the proliferation of smartphones and their increasingly sophisticated sensing capabilities had the potential to improve health care outcomes for millions of people around the globe. He and his students began thinking about how they could employ these on-board sensors — such as the phone’s camera, microphone, accelerometer, and gyroscope — to augment traditional in-person care by enabling early detection and intervention. They also saw an opportunity to empower people to monitor their health on an ongoing basis, without the need for repeated trips to a clinic or access to specialized equipment, with the help of a device they already carry around with them.

Working with clinicians at UW Medicine, Seattle Children’s and others, Patel and his team developed apps for assessing lung function in people with respiratory illnesses, detecting jaundice in babies and adults, measuring blood hemoglobin in people with anemia — to name only a few. Patel and his collaborators started a company, Senosis Health, to commercialize their research. After Senosis was acquired by Google, Patel began splitting his time between the UW and the company in order to lead the latter’s mobile health efforts. Patel and the Google Fit team have since released tools for measuring heart rate and respiratory rate to permit users to monitor their general health and wellness with the aid of their smartphone camera that was based in part on research originating in his UW lab.

Following the emergence of SARS-CoV-2 early last year, Patel and his lab pivoted to applying what they had learned from their work on those earlier apps to focus on tools that could aid in the pandemic response. For example, Patel and his students have been working on smartphone-based tools for monitoring symptoms such as cough and developed a system to enable contactless measurement of a person’s vital signs via online video. To aid in the community-level response, he also collaborated on the creation of RDTScan to support accurate interpretation of rapid diagnostic test results at the point of capture with the help of a smartphone and demonstrated how air filtration systems on public transit could be used as passive sensing systems to detect viral spread.

Patel’s efforts to advance mobile health sensing were a natural progression from his visionary work on low-power sensing that stretches back to his student days at Georgia Tech. His first foray into the technology was as an undergraduate working on the Aware Home, a demonstration project that sought to imagine the connected home of the future. After earning his bachelor’s, Patel remained in Atlanta to pursue his doctorate, during which time he developed a system for measuring residential energy and water consumption by individual appliances and fixtures from a single point in the home — research that Patel continued to refine and expand upon following his arrival at the UW. He and his Georgia Tech collaborators started a company, Zensi, to commercialize that work which was subsequently acquired by Belkin. 

Next, Patel and his students zoomed out from looking at individual appliances to monitoring the entire home via an ultra-low-power sensing system known as SNUPI, short for Sensor Nodes Utilizing Powerline Infrastructure. SNUPI consisted of a network of low-power sensors that transmitted data about a building’s condition — for example, increased moisture level in the walls — via the structure’s electrical circuit. The system was designed to function for decades without having to replace the batteries. Patel and his team created another spinout company, SNUPI Technologies, to commercialize a residential whole-home hazards monitoring platform under the name of WallyHome that was later acquired by Sears. 

Through the years, Patel and his students have also advanced innovations in motion tracking, object detection, wearable technologies, hyperspectral imaging, and more. Throughout his career, he has earned more than two dozen Best Paper Awards and multiple “test of time” awards at the field’s preeminent conferences focused on ubiquitous computing, mobile computing, pervasive computing and human-computer interaction.

Patel’s induction into his alma mater’s Hall of Fame and the Business Insider recognition are the latest in a string of accolades recognizing the wide-ranging impact of his work. Previously, he was named a Fellow of the Association for Computing Machinery and received the organization’s ACM Prize in Computing for mid-career contributions to the field. Patel is also a past recipient of a MacArthur Foundation “Genius” Award and a Presidential Early Career Award for Scientists and Engineers (PECASE).

Read the Georgia Tech College of Computing Hall of Fame citation here, Patel’s “30 under 40” profile in Business Insider here, and a related article on his work on health care technologies at Google here (paywall).

Congratulations, Shwetak — times two!

Photo credit: Matt Hagen

Read more →

“They were very unsure what to do with us”: How Golden Goose Award-winning researchers at UW and UCSD put the brakes on automobile cybersecurity threats and transformed an industry

Collage of portraits of Stephen Checkoway, Karl Koscher, Stefan Savage and Tadayoshi Kohno
UW and UCSD Golden Goose Award recipients (clockwise from top left): Stephen Checkoway, Karl Koscher, Stefan Savage and Tadayoshi Kohno

In 2010 and 2011, a team of researchers led by Allen School professor Tadayoshi Kohno and Allen School alumnus and University of California San Diego professor Stefan Savage (Ph.D., ‘02) published a pair of papers detailing how they were able to hack into a couple of Chevrolet Impalas and take command of a range of functions, from operating the windshield wipers to applying — or even denying — the brakes. A decade later, Kohno, Savage, and University of Washington alumni Karl Koscher (Ph.D., ‘14), now a research scientist in the Allen School’s Security & Privacy Research Lab, and Stephen Checkoway (B.S., ‘05), a UCSD Ph.D. student at the time who is now a faculty member at Oberlin College, have received the Golden Goose Award from the American Association for the Advancement of Science for demonstrating “how scientific advances resulting from foundational research can help respond to national and global challenges, often in unforeseen ways.”

“More than 10 years ago, we saw that devices in our world were becoming incredibly computerized, and we wanted to understand what the risks might be if they continued to evolve without thought toward security and privacy,” explained Kohno in a UW News release.

Achieving that understanding would go on to have significant real-world impact, influencing “how products are built and how policies are written,” noted Savage. It would also transform not just the automobile manufacturing landscape, but the computer security research landscape as well. 

“The entire automotive security industry grew from this effort,” recalled Kohno. “And I imagine that neighboring industries saw what happened here and didn’t want something similar happening to them.”

“What happened here” was that Kohno and his colleagues demonstrated how a motor vehicle’s computerized systems could be vulnerable to attackers, theoretically endangering the car’s occupants and those who share the road with them. The quartet was aided and abetted by collaborators that included, on the Allen School side, then-student and current professor Franziska Roesner (Ph.D., ‘14), fellow student Alexei Czeskis (Ph.D., ‘13), and professor Shwetak Patel; on the UCSD side, they were joined by postdoc Damon McCoy, master’s student Danny Anderson, professor Hovav Shacham, and the late researcher Brian Kantor.

This “dream team,” as Kohno describes it, set out to reverse-engineer the various vehicle components. The goal was to figure out how they communicated with each other so that they could use that to gain access to the systems that control the vehicle’s functions. The researchers published two papers in rapid succession detailing their findings; the first established how a car’s internal systems were vulnerable to compromise, while the follow-up explored the external attack surface of the vehicle by demonstrating how an attacker could infiltrate and control those systems remotely. The team presented the former at the 2010 IEEE Symposium on Security & Privacy and the latter at the 2011 USENIX Security Symposium.

In a way, Savage recalled, the researchers’ ignorance about how the vehicle’s systems were actually designed to work ended up working to the team’s advantage; it enabled them to approach their task without any preconceptions of what should happen. An example is the brake controller, which they broke into via a technique known as black-box testing or “fuzzing.” As the label suggests, these efforts involved less precision and more “throwing stuff at it,” according to Savage, to see what would stick. The results were enough to stop anyone in their tracks — including the technical experts at GM.

“We figured out ways to put the brake controller into this test mode,” Koscher explained to UW News. “And in the test mode, we found we could either leak the brake system pressure to prevent the brakes from working or keep the system fully pressurized so that it slams on the brakes.”

As the senior Ph.D.s on the project, Koscher and Checkoway spearheaded that discovery, which involved calling into the car’s OnStar unit and instructing it to download and install remote command and control software that they had written. With that in place, they were able to compel the system to download the software that would enable them to remotely control the brakes from a laptop — as demonstrated later in a famous “60 Minutes” segment in which the team surprised correspondent Leslie Stahl by bringing the car to a complete stop while she was behind the wheel.

While that made for good television, what is most gratifying for the researchers are the industry and regulatory frameworks that grew out of their discovery. 

For example, GM — along with other manufacturers — hired an entire security team as a direct result of the UW and UCSD research; likewise, the National Highway Traffic Safety Administration (NHTSA) — which previously had no one on staff with computer security expertise and “were very unsure what to do with us,” according to Savage — wound up creating an entire unit devoted to cybersecurity, complete with its own testing lab. In other positive changes, the Society of Automotive Engineers — later renamed SAE International — established a set of security standards that all automobile manufacturers adhere to, and the industry created the Auto-ISAC, a national Information Sharing and Analysis Center, to enhance vehicle cybersecurity and address emerging threats.

The team’s work also paved the way for new research outside of the automotive industry. For example, its results inspired the U.S. Defense Advanced Research Projects Agency (DARPA) to create its HACMS project, short for High-Assurance Cyber Military Systems, to examine the security of cyber-physical systems. And that was just the start.

“My gut tells me that the attention directed at this project helped to build up expertise in this embedded systems realm,” observed Koscher. “What was initially focused on automotive security was then applied to other industries, such as medical devices.”

The project also served to highlight the advantages of working as part of a larger team, as Checkoway discovered to his delight. While various members of the group may have approached a problem from different angles, they would often meet in the middle to come up with a solution.

“This was an extremely collaborative effort,” Checkoway explained last year. “No task was performed by an individual researcher alone. I believe our close collaboration was the key to our success.”

At the time the researchers quietly revealed their results to GM, they couldn’t be sure such a happy outcome was a foregone conclusion. At first, the company representatives didn’t believe they could do some of the things they had done — or how they could have possibly done them. But the team’s non-adversarial approach, in which they opted to walk company representatives through their process and findings while refraining from naming the manufacturer publicly, went a long way toward steering the conversations in a positive direction.

“As academics, we have the opportunity to approach the dialogue around vulnerabilities without really having a stake in the game,” explained Kohno. “We’re not selling vulnerabilities, we’re not selling a product to patch vulnerabilities, and we aren’t a competing manufacturer. So we discovered something, and once we had the results, we wanted to figure out, how can we use this knowledge to make the world a better place?”

The team is quick to credit the federal government for driving investment in a project for which they didn’t have a precise destination in mind when they started. According to Savage, the National Science Foundation’s willingness to back a project that was not guaranteed to pan out was key to enabling them to identify these latent security risks. “We’re extremely grateful to NSF for having flexibility to fund this work that was so speculative and off the beaten path,” Savage said.

Two men wearing masks standing next to car, one is typing on laptop set on car roof
Checkoway (left) and Koscher reunite with Emma the Impala in the UW’s Central Garage Mark Stone/University of Washington

It is just the kind of work that the Golden Goose Award was created to recognize. In answer to the late U.S. Senator William Proxmire’s “Golden Fleece Award” ridiculing federal investment that he deemed wasteful, U.S. Representative Jim Cooper conceived of the Golden Goose Award to honor “the tremendous human and economic benefits of federally funded research by highlighting examples of seemingly obscure studies that have led to major breakthroughs and resulted in significant societal impact.”

For Kohno, that impact and this most recent recognition — the team previously earned a Test of Time Award from the IEEE Computer Society Technical Committee on Security and Privacy — are motivation enough to explore where the next security risk may come from. 

“The question that I have now is, as security researchers, what should we be investigating today, such that we have the same impact in the next 10 years?”

The team was formally honored in a virtual ceremony hosted by the AAAS last week. Read the feature story on the UW and UCSD team here, the UW News release here and a related story by Oberlin College here. Learn about all of the 2021 Golden Goose honorees here.

Read more →

A snapshot of the future: Computing goes molecular with DNA-based similarity search from University of Washington and Microsoft researchers

Collection of 20 close-up photos of cats in varying poses, arranged in a grid
Forget social media memes — searching for cat photos is serious business for members of the Molecular Information Systems Lab.

Picture this: You are a researcher in the Molecular Information Systems Lab housed at the Paul G. Allen School. You and your labmates have developed a system for storing and retrieving digital images in synthetic DNA to demonstrate how this extremely dense and durable medium might be used to preserve the world’s growing trove of data. And you have a particular fondness for cats.

If you wanted to sift through photos of frisky felines — and really, what better way to spend an afternoon — how would you pick out the relevant files floating around in your test-tube database without having to sequence the entire pool?

In a paper published recently in the journal Nature Communications, a team of MISL scientists presented the first technique for performing content-based similarity search among digital image files stored in DNA molecules. The approach is akin to that of a modern search engine, albeit in a much smaller form factor than your average server farm and with the potential to be much more energy efficient.

”Content-based search enables us to type a word or phrase into a hundred-page document and be taken to the exact page it appears, or upload a photo of a daisy and get flower images in return,” said co-author Yuan-Jyue Chen, senior researcher at Microsoft and an affiliate professor at the Allen School. “We don’t know the specific page numbers or files we’re looking for, so that computation saves us the trouble of reading the entire document or scrolling through every photo on the internet. Our team took that same idea and applied it to data stored in molecular form.”

Files, whether stored in digital or molecular form, make use of a process known in database parlance as key-based retrieval. In an electronic database, it is typically a file name; in a molecular database, it is a unique sequence, reminiscent of a barcode, that is encoded in the snippets of DNA associated with a particular file. Items with this barcode can be amplified via polymerase chain reaction (PCR) to reassemble a file in its entirety, since a single digital file might be split among hundreds — possibly even thousands — of DNA oligonucleotides, depending on its total size. Generally speaking, key-based retrieval works great when you know the contents of the files and can pick out which ones you want to retrieve; if you don’t and your data is stored as the As, Ts, Cs and Gs of DNA instead of 0s and 1s, the entire database would have to be sequenced in order to perform a content-based search.

Group shot of team members standing in front of metal railing in light-filled atrium constructed of glass, concrete and brick
Left to right: Luis Ceze, Yuan-Jyue Chen, Callista Bee, Karin Strauss, David Ward and Aaron Liu. Tara Brown Photography

To move beyond the limitations of key-based retrieval, the researchers leveraged DNA’s natural hybridization behavior along with machine learning techniques to enable similarity search to be performed on the stored data. In a conventional digital database, similarity search relies on a set of feature vectors that are stored separately from the original data. When a search is executed, its results point to the location of each data file associated with a particular feature vector. For the molecular version, the MISL team developed an image-to-sequence encoding scheme that employs a convolutional neural network to designate image feature vectors as “similar” or “not similar” and then maps them to DNA sequences that will predictably hybridize — or bind — with the reverse complement of a query feature vector processed by the same neural network during execution of a search. The technique can be applied to new images not seen during training, and the entire process is easily extended to other types of data such as videos and text.

The researchers created an experimental database by running a collection of 1.6 million images through their encoder, which converted them to DNA sequences incorporating their feature vectors, and tacked on a unique barcode identifier for each file. They then performed a similarity search for three photos — including one of a tuxedo cat named Janelle — using the reverse complement of each query image’s encoded feature sequence against a sample of the database. After filtering out the hybridized target/query pairs for high-throughput sequencing, they found the most frequently sequenced oligos did, indeed, corresponded to images in the database that were visually similar to the query images.

The team found that its molecular-based approach was comparable to that of in silico algorithms representing the state of the art in similarity search. Unlike those algorithms, however, the team points out that DNA-based search has the potential to scale to significantly larger databases without a correspondingly significant increase in processing time and energy consumption due to its inherently parallel nature. In this way, researchers have barely scratched the surface of what DNA computing can do.

Portraits of Lee Organick, Melissa Queen and Georg Seelig
Left to right: Lee Organick, Melissa Queen and Georg Seelig. Tara Brown Photography

“As DNA data storage is made more practical by advances in fast and low-cost synthesis, sequencing and automation, the ability to perform computation within these databases will pave the way for hybrid molecular-electronic computer systems. Starting with similarity search is exciting because that is a popular primitive in machine learning systems, which are quickly becoming pervasive,” Allen School professor and co-corresponding author Luis Ceze said.

In addition to Ceze and Chen, contributors to the paper include lead author and Allen School alumna Callista Bee (Ph.D., ‘20), Ph.D. students Melissa Queen and Lee Organick, former research scientist Xiaomeng (Aaron) Liu, lab manager David Ward, Allen School and UW Electrical & Computer Engineering professor Georg Seelig, and co-corresponding author Karin Strauss, an affiliate professor in the Allen School and senior principal research manager at Microsoft. Bee initially presented the team’s proof of concept and precursor to this work at the 24th International Conference on DNA Computing and Molecular Programming (DNA 24), for which she earned a Best Student Paper Award.

Read the team’s latest paper, “Molecular-level similarity search brings computing to DNA data storage,” in Nature Communications.

Read more →

Allen School’s Dhruv Jain wins Microsoft Research Dissertation Grant for his work leveraging HCI and AI to advance sound accessibility

Dhruv Jain smiling in front of glass and wood cabinet

Allen School Ph.D. student Dhruv (DJ) Jain has received a Microsoft Research Dissertation Grant for his work on “Sound Sensing and Feedback Techniques for Deaf and Hard of Hearing Users.” This highly selective grant aims to increase diversity in computing by supporting doctoral students who are underrepresented in the field to “cross the finish line” during the final stages of their dissertation research. 

Jain, who is co-advised by Allen School professor Jon Froehlich and Human Centered Design & Engineering professor and Allen School adjunct professor Leah Findlater, works in the Makeability Lab to advance sound accessibility by designing, building and deploying systems that leverage human computer interaction (HCI) and artificial intelligence (AI). His primary aim is to help people who are d/Deaf and hard of hearing (DHH) to receive important and customized sound feedback. The dissertation grant will support Jain’s continuing work on the design and evaluation of three of these systems. 

One of his projects, HomeSound, is a smart home system that senses and visualizes sound activity like the beeping of a microwave, blaring of a smoke alarm or barking of a dog in different rooms of a home. It consists of a microphone and visual display, which could be either a screen or a smartwatch, with several devices installed throughout the premises. Another system, SoundWatch, is an app that provides always-available sound feedback on smartwatches. When the app picks up a nearby sound like a car honking, a bird chirping or someone hammering, it sends the user a notification along with information about the sound. Jain also has contributed to the development of HoloSound, an augmented reality head-mounted display system that uses deep learning to classify and visualize the identity and location of sounds in addition to providing speech transcription of nearby conversations. All three projects are currently being deployed and tested with DHH users.

“Dhruv is a dedicated researcher who draws on his own unique life experiences to design and build interactive systems for people who are deaf or hard of hearing,” Froehlich said. “DJ cares not just about academic results and solving hard technical problems but in pushing towards deployable solutions with real-world impact. SoundWatch is a great example: DJ helped lead a team in building a real-time sound recognizer in the lab that they then translated to an on-watch system deployed on the Google Play Store. Thus far, it’s been downloaded over 500 times.”

Over the course of his research, Jain has published 20 papers at top HCI venues such as the Association for Computing Machinery’s Conference on Human Factors in Computing Systems (CHI), Symposium on User Interface Software and Technology (UIST), Designing Interactive Systems Conference (DIS) and Conference on Computers and Accessibility (ASSETS). His work has received two Best Paper Awards, three Best Paper Honorable Mentions and one Best Artifact Award. 

Learn more about the 2021 grant recipients here.

Congratulations, DJ!

Read more →

CAREER Award-winning faculty at the Allen School advance leadership and innovation in software testing, machine learning equity, and natural language understanding

NSF logo

How can we endow artificial intelligence with the ability to comprehend and draw knowledge from the immense and varied trove of online documents? Is there a way to make testing and debugging of software that pervades our increasingly technology-dependent world more efficient and robust? Speaking of pervasiveness, as technologies like machine learning are more embedded into our society, how can we be sure that these systems reflect and serve different populations equitably? 

Those are the questions that Allen School professors Hannaneh Hajishirzi, René Just and Jamie Morgenstern are grappling with in their latest research. And while each focuses on a different area, they all have at least two things in common: Their commitment to integrating research and education earned them National Science Foundation CAREER Awards recognizing junior faculty who exemplify the role of teacher-scholars, and their contributions are helping to cement the Allen School’s leadership in both core and emerging areas of computing.

Hannaneh Hajishirzi: Knowledge-rich neural text comprehension and reasoning

Hannaneh Hajishirzi

Hajishirzi joined the Allen School faculty full-time in 2018 and is the director of the H2Lab, where she focuses on finding and addressing foundational problems in artificial intelligence (AI) and natural language processing (NLP). She is also a Fellow at the Allen Institute for AI. In her CAREER award-winning work, Hajishirzi aims to improve textual comprehension and reasoning using AI capabilities and deep learning algorithms to help applications better understand text and draw logical conclusions. For example,  her research seeks to enable AI systems to answer questions such as “What percentage of Washington state’s budget has been spent on education over the last 20 years?” or verify claims like “Baricitnib prevents 2019-COV from infecting AT2 lung cells.” To do so, the AI system needs to comprehend the claim, find data from reliable sources like scientific articles, and perform reasoning skills to integrate explicit and implicit evidence. Hajishirzi will integrate AI capabilities into deep learning algorithms to understand and reason about textual comprehension to devise hybrid, interpretable algorithms that understand and reason about textual knowledge across varied formats and styles. It will also generalize to emerging domains with scarce training data and operate efficiently under resource limitations.

“Effectively unlimited quantities of ever-changing knowledge are available online in diverse styles, such as news vs. science text and formats — knowledge bases, financial reports and textual documents,” Hajishirzi said. “The content and language style in these domains might evolve over time; for example, scientific articles about the Covid-19 pandemic may use different vocabulary and style as scientists learn more about the virus and communicate findings to different audiences. It is really important to build AI systems that make sense of this enormous amount of knowledge. As a result, textual comprehension is a fundamental problem in AI and NLP.”

René Just: Toward effective, predictable, and consistent software testing

René Just

Since joining the Allen School faculty in 2018, Just has focused on software engineering and data science, in particular static and dynamic program analysis, empirical software engineering and applied statistics and machine learning. Through his CAREER-winning research,  Just is working to increase the quality of the software that powers modern technology and to improve software development more generally with a framework and methodology for systematic software testing. By developing more effective and consistent software testing approaches, Just will provide developers with concrete test goals quantifying the degree to which these test goals are representative of the defects that were experienced during development in the past. By developing more effective and consistent software testing, his method will be able to assess test goals to check for bugs that were experienced during the development of the software.

“Given that software affects nearly every aspect of our daily lives, its defects have serious implications, and significant advances in software quality benefit every corner of society,” Just said. “This is the main motivation for my work on software testing, which aims to prevent fatal software incidents, million-dollar defects, and a lot of frustration caused by software that simply isn’t working as expected.”

Jamie Morgenstern: Strategic and equity considerations in machine learning

Jamie Morgenstern

Morgenstern, who joined the Allen School faculty in 2019, studies the social impact of machine learning (ML) and the impact of social behavior because of ML. In her CAREER award-winning work, Morgenstern is researching the performance of machine learning systems that are trained and tested on heterogeneous and strategically generated data and make both predictions and decisions. This work will help guarantee high-quality predictions in domains with varied data sources. It will also help ensure that insights from ML will apply to diverse populations rather than just the majority. Morgenstern’s work will transform the way both researchers and practitioners reason about human-centric ML models. Additionally, it will shed light on interesting technical questions about how optimizing for different measures of performance affects minority communities. 

“This will provide guarantees on the future performance of systems built on human-generated historical data, even when those people will continue to interact with the system,” Morgenstern said. “My research will inform the design of learning algorithms trained and deployed on human-generated data, in domains such as commerce, medicine, risk modeling, and social benefit allocation.”

Morgenstern, Just and Hajishirzi are the most recent Allen School faculty members to advance leading-edge research with support from the NSF CAREER program. A total of 64 faculty members have received one of these prestigious awards or their predecessor, the Presidential/NSF Young Investigator Award, during their time at the Allen School.

Read more →

University of Washington and Microsoft researchers develop “nanopore-tal” enabling cells to talk to computers

Two people wearing masks and gloves in a lab. One is sitting at a table piping cell culture onto a small rectangular device connected to a laptop, surrounded by various lab supplies. Second person is standing behind first person observing.
MISL researcher Nicolas Cardozo pipes cell cultures containing NanoporeTERs onto a portable MinION nanopore sensing device for processing as professor Jeff Nivala looks on. Dennis Wise/University of Washington

Genetically encoded reporter proteins have been a mainstay of biotechnology research, allowing scientists to track gene expression, understand intracellular processes and debug engineered genetic circuits. But conventional reporting schemes that rely on fluorescence and other optical approaches come with practical limitations that could cast a shadow over the field’s future progress. Now, thanks to a team of researchers at the University of Washington and Microsoft, scientists are about to see reporter proteins in a whole new light. 

In a paper published today in the journal Nature Biotechnology, members of the Molecular Information Systems Laboratory housed at the UW’s Paul G. Allen School of Computer Science & Engineering introduce a new class of reporter proteins that can be directly read by a commercially available nanopore sensing device. The new system ― dubbed “Nanopore-addressable protein Tags Engineered as Reporters,” also known as NanoporeTERs or NTERs for short ― can perform multiplexed detection of protein expression levels from bacterial and human cell cultures far beyond the capacity of existing techniques. 

You could say the new system offers a “nanopore-tal” into what is happening inside these complex biological systems where, up until this point, scientists have largely been operating in the dark.

“NanoporeTERs offer a new and richer lexicon for engineered cells to express themselves and shed new light on the factors they are designed to track. They can tell us a lot more about what is happening in their environment all at once,” said co-lead author Nicolas Cardozo, a graduate student in the UW’s molecular engineering Ph.D. program. “We’re essentially making it possible for these cells to ‘talk’ to computers about what’s happening in their surroundings at a new level of detail, scale and efficiency that will enable deeper analysis than what we could do before.” 

Laptop screen showing squiggly lines of various colors stacked on top of each other, representing signals from a nanopore sensing device
Raw nanopore signals streaming from the MinION device, which contains an array of hundreds of nanopore sensors; each color represents data from an individual nanopore. The team uses machine learning to interpret these signals as NanoporeTERs barcodes. Dennis Wise/University of Washington

Conventional methods that employ optical reporter proteins, such as green fluorescent protein (GFP), are limited in the number of distinct genetic outputs that they can track simultaneously due to their overlapping spectral properties. For example, it’s difficult to distinguish between more than three different fluorescent protein colors, limiting multiplexed reporting to a maximum of three outputs. In contrast, NTERs were designed to carry distinct protein “barcodes” composed of strings of amino acids that, when used in combination, enable a degree of multiplexing approaching an order of magnitude more. These synthetic proteins are secreted outside of the cell into the surrounding environment, where they are collected and directly analyzed using a commercially available nanopore array — in this case, the Oxford Nanopore Technologies MinION device. To make nanopore analysis possible, the NTER proteins were engineered with charged “tails” that get pulled into the tiny nanopore sensors by an electric field. Machine learning is then used to classify their electrical signals in order to determine the output levels of each NTER barcode.

“This is a fundamentally new interface between cells and computers,” explained Allen School research professor and corresponding author Jeff Nivala. “One analogy I like to make is that fluorescent protein reporters are like lighthouses, and NanoporeTERs are like messages in a bottle. Lighthouses are really useful for communicating a physical location, as you can literally see where the signal is coming from, but it’s hard to pack more information into that kind of signal. A message in a bottle, on the other hand, can pack a lot of information into a very small vessel, and you can send many of them off to another location to be read. You might lose sight of the precise physical location where the messages were sent, but for many applications that’s not going to be an issue.”

In developing this new, more expressive vessel, Nivala and his colleagues eschewed time-consuming sample preparation or the need for other specialized laboratory equipment to minimize both latency and cost. The NTERs scheme is also highly extensible. As a proof of concept, the team developed a library of more than 20 distinct tags; according to co-lead author Karen Zhang, the potential is significantly greater.

Four people standing posed against a glass and metal railing in light-filled atrium.
Co-authors of the Nature Biotechnology paper (left to right): Karen Zhang, Nicolas Cardozo, Kathryn Doroschak and Jeff Nivala. Not pictured: Aerilynn Nguyen, Zoheb Siddiqui, Nicholas Bogard, Karin Strauss and Luis Ceze. Tara Brown Photography

“We are currently working to scale up the number of NTERs to hundreds, thousands, maybe even millions more,” Zhang, who graduated this year from the UW with bachelor’s degrees in biochemistry and microbiology, explained. “The more we have, the more things we can track. We’re particularly excited about the potential in single-cell proteomics, but this could also be a game-changer in terms of our ability to do multiplexed biosensing to diagnose disease and even target therapeutics to specific areas inside the body. And debugging complicated genetic circuit designs would become a whole lot easier and much less time consuming if we could measure the performance of all the components in parallel instead of by trial and error.”

MISL researchers have made novel use of the ONT MinION device before. Allen School alumna Kathryn Doroschak (Ph.D., ‘21), one of the lead co-authors of this paper, was also involved in an earlier project in which she and her teammates developed a molecular tagging system to replace conventional inventory control methods. That system relied on barcodes comprising synthetic strands of DNA that could be decoded on demand using the portable ONT reader. This time, she and her colleagues went a step further in demonstrating how versatile such devices can be.

“This is the first paper to show how a commercial nanopore sensor device can be repurposed for applications other than the DNA and RNA sequencing for which they were originally designed,” explained Doroschak. “This is exciting as a precursor for nanopore technology becoming more accessible and ubiquitous in the future. You can already plug a nanopore device into your cell phone; I could envision someday having a choice of ‘molecular apps’ that will be relatively inexpensive and widely available outside of traditional genomics.”

Additional co-authors of the paper include research assistants Aerilynn Nguyen and Zoheb Siddiqui, former postdoc Nicholas Bogard, Allen School affiliate professor Karin Strauss, senior principal research manager at Microsoft; and Allen School professor Luis Ceze.

Read the paper, “Multiplexed direct detection of barcoded protein reporters on a nanopore array,” in Nature Biotechnology.

Editor’s note: Team photo was taken pre-pandemic.

Read more →

Kyle Johnson wins 2021 Generation Google Scholarship

Kyle Johnson

Kyle Johnson, a Ph.D. student working with professor Shyam Gollakota in the Allen School’s Network and Mobile Systems Lab, received a Generation Google Scholarship for his academic performance, leadership and commitment to diversity, equity and inclusion. The company created the scholarship to help students pursuing computer science degrees excel in technology and become leaders in the field. 

As a researcher, Johnson aims to create battery-less micro robots designed to operate autonomously for prolonged periods of time, implementable as the sensory notes in a swarm algorithm. To achieve this, he plans to leverage the properties of structures like leaf-out origami to create origami robots. These have potential applications in many environments requiring both low-power and small-scale devices, like in the deployment of space rovers for interplanetary exploration. 

“As a first year graduate student, Kyle is leading embedded systems and robotic research that span aerospace, electrical and mechanical engineering and computer science,” said Gollakota. “The miniaturized robotic systems he is building, if successful, are creative and more importantly useful for achieving wireless sensor deployment at a scale that has not been possible before.”

In addition to his research, Johnson has been a leader in working towards improving diversity, equity and inclusion in academia and technology fields. For the past three years, Johnson has taught middle and high school students how to code and hardwire a multitude of different sensors and devices, demonstrating to students that they can apply the technical concepts that they learn in class towards solving real-world problems.

In the summer of 2019, Johnson traveled to Cape Town, South Africa to study racism, education and development. There, he learned from high school math and physics teachers about creating lesson plans and teaching to diverse students before eventually teaching some of those classes himself. At the University of South Africa, Johnson shared his experiences during the “Trauma, Educational Exclusions and Survival: Examining Global Student Experience & Resilience” workshop at an academic development symposium and encouraged those at the university to be more engaged in public schools to better prepare students for higher education. 

Upon his return to the United States, Johnson co-founded the student group A Vision for Electronic Literary & Access (AVELA) to provide more opportunities for underrepresented students to pursue their interests at a university level. Johnson and the other AVELA members accomplish this by representing the populations that they aim to support while leading workshops, camps and other forms of community outreach. AVELA has partnered with Seattle Public Schools, the Kent School District, the National Society of Black Engineers, Seattle MESA, The Urban League of Metropolitan Seattle, InspireOne, and the city of Seattle to teach engineering workshops and create lesson plans catered towards aspiring, underrepresented students. The organization is in the process of earning a 501(c) status; as a nonprofit, AVELA can apply for grants and additional funding.

Over the past year, Johnson has worked to help change policies and job descriptions at the University of Washington to make them more equitable and created tools for students to more quickly report racism, sexism and other discriminatory actions. He also co-founded the Black Student Graduate Association (BSGA).

““The BGSA focuses on giving UW’s Black graduate students the space to relax, network and share experiences,” Johnson said. “As the 100-member organization expands, I hope more Black graduate students will find solace in a community of their peers.”

In addition to his Google Scholarship, Johnson is also a recipient of the National Science Foundation Graduate Research Fellowship, the National GEM Consortium Fellowship and the Washington Research Foundation Scholarship. And he is a LEAP Fellow – the LEAP Alliance is a collaboration among Berkeley, CMU, Cornell, Georgia Tech, Harvard, Illinois, MIT, Princeton, Stanford, Texas, and Washington focused on diversifying LEAdership in the Professoriate.

Congratulations, Kyle! 

Read more →

« Newer PostsOlder Posts »