Overview
The Allen School has a long and proud tradition of applying computing innovation to address societal challenges and improve people’s lives — from portable technologies for making the web accessible to people who are blind or low vision, to a suite of mobile data collection tools deployed around the world in support of public health, sustainable agriculture, wildlife conservation, and more. Today, our researchers continue to build on this tradition while opening up new avenues of exploration for advancing computer science for good. Below is a sampling of our research that aims to solve problems with broad impact for people, society and our planet while expanding the frontiers of our field.
Computing for Health
Allen School researchers, in collaboration with health scientists and clinicians, are applying computing innovation to transform medical science and patient care — from leveraging advances in artificial intelligence and machine learning to cultivate a deeper understanding of diseases such as Alzheimer’s and cancer, to devising new approaches to diagnosis and treatment that harness the sensing capabilities of smartphones and other smart devices. Our researchers are also applying data science techniques to gain insights into how various factors impact people’s well-being at a larger scale. And when COVID-19 emerged as a public health threat, we rapidly turned our expertise to aiding the public health response — work which will have an enduring impact beyond the current pandemic.
Sample Projects & Papers
- Deep neural networks for understanding the role of gene expression in Alzheimer's disease
- Interpretable, real-time predictions of patient risk factors to prevent hypoxemia during surgery
- An automated approach for more accurate breast cancer diagnosis using machine learning and computer vision
- A framework for designing next-generation mRNA therapeutics
- Tumor stratification across cancer types via adversarial learning and consensus clustering
- Contactless measurement of vital signs over video to support telehealth
- At-home screening for pediatric ear infections using a smartphone
- A wearable automatic naloxone injector system for rapid intervention in the event of an opioid overdose
- Online testing to support people with cognitive and mental disabilities
- Design principles for privacy-preserving mobile contact tracing apps
- An analysis of the impact of AI shortcut learning on COVID-19 diagnosis from lung x-rays
- Accurate interpretation of rapid diagnostic test results via smartphone
- Insights into how people's ability to identify with greater humanity shapes their response to humanitarian crises
Recent News
Smartphone system measures clotting time of users' blood samples
Hear, hear: Sound Life Sciences wins FDA OK for app that uses sonar to measure breathing
Wearable device can spot, reverse opioid overdoses, researchers say
'Lazy' AI: UW researchers find that tech can misdiagnose COVID-19 by taking shortcuts
Contributing Faculty
- Tim Althoff
- James Fogarty
- Shyam Gollakota
- Vikram Iyer
- Su-In Lee
- Jennifer Mankoff
- Sara Mostafavi
- Shwetak Patel
- Rajesh Rao
- Katharina Reinecke
- Larry Ruzzo
- Georg Seelig
- Linda Shapiro
- Joshua Smith
- Stefano Tessaro
- Sheng Wang
Computing for Sustainability
Allen School researchers are taking inspiration from nature and developing innovative solutions to address the world’s sustainability challenges while reducing the carbon footprint of computing itself. Some of our work focuses on untethering computation from traditional power sources to support a variety of real-world applications, from smart agriculture, to green buildings, to disaster response. Our researchers are also reimagining how we manage the world’s rapidly expanding trove of digital data by developing novel storage capabilities using synthetic DNA as a potential alternative to power-hungry datacenters — and designing other systems and hardware that will optimize performance while minimizing environmental impact across a variety of domains, from artificial intelligence to manufacturing.
In 2022, the Allen School launched Computing for the Environment (CS4Env) in partnership with the University of Washington's Department of Civil & Environmental Engineering, College of the Environment, College of Engineering and eScience Institute. The goal of the initiative is to support novel collaborations and cultivate a community of researchers across the Environmental Sciences and Computer Science & Engineering. The program awarded its first round of competitive research grants to 12 cross-disciplinary projects spanning air quality monitoring, wildfire forecasting, climate modeling, wildlife tracking, wastewater epidemiology, and more.
Sample Projects & Papers
- Long-distance wireless communication and computation through long-range backscatter
- A stick-on capacitive sensor for harvesting energy from the stray electric field around power lines
- The world’s first functioning battery-free cellphone
- Single-point sensors for monitoring home energy usage at the appliance level and water usage at the fixture level
- Solar-powered wireless sensors that can be dispersed on the wind like dandelion seeds
- A wireless steerable vision system for insects and insect-sized robots
- Lightweight battery-powered sensors that can be airdropped from small drones and insects
- A flying wireless platform for sensing and communication carried by insects
- A system for transforming wireless transmissions from one technology to another, over the air, to enable internet-connected implantable devices
- Molecular-level similarity search for digital image files stored in synthetic DNA
- The first demonstration of an end-to-end system for automating DNA-based digital data storage
- A molecular tagging system for physical objects
- A "DNA domino" architecture based on spatial organization to enable fast and modular DNA computing
- Coming soon!
Recent News
Dandelion-inspired sensors float on the wind
‘World’s lowest power wireless chip’: UW spinout Jeeva unveils new sensor data streaming tech
We built a 2020 time capsule out of synthetic DNA—here's how
DNA-based molecular tagging system could replace printed barcodes
UW researcher put tiny tracking technology on giant hornets to help state deal with murderous pest
‘Carpentry Compiler’ turns 3D models to instructions on how to build them
Contributing Faculty
- Luis Ceze
- Yejin Choi
- James Fogarty
- Jon Froehlich
- Shyam Gollakota
- Hannaneh Hajishirzi
- Arvind Krishnamurthy
- Vikram Iyer
- Jennifer Mankoff
- Jeff Nivala
- Shwetak Patel
- Simon Peter
- Georg Seelig
- Joshua Smith
- Noah Smith
- Michael Taylor
- Luke Zettlemoyer
Computing for Society
It’s a small world after all, as the famous song goes — and it’s only getting smaller thanks to our increasing ability to connect instantly with people around the globe and to access, translate, and disseminate content online. However, this easy connectivity has a downside; it can be misused to stoke political and social divides, rapidly spread misinformation, target vulnerable groups, and exploit weaknesses in privacy and security. The positive impacts of technology can also be unevenly distributed, particularly in low-resource communities. This is where Allen School researchers come in, analyzing and mitigating adverse impacts of emerging technologies and giving users the tools they need to protect themselves and others while enjoying the benefits of a more connected world through computing.
Sample Projects & Papers
- Evaluation and prevention of toxic degeneration in pre-trained neural language models
- Intersectional analysis of social biases spanning gender and race in online narrative text
- An exploration of AI's capacity for modeling ethics and social awareness
- A model framework for regulating the use of facial recognition technologies
- Techniques for identifying and mitigating racial bias in automated hate-speech detection tools
- Best practices and tools for safeguarding user privacy and security in augmented reality
- An analysis of security vulnerabilities in genetic genealogy services that could put individuals' personal genetic information at risk
- Effective and usable tools for protecting sensitive communications between journalists and sources
- An evaluation of computer security risks in modern automobiles
Recent News
Can computers learn common sense?
UW and UC San Diego researchers honored for their work discovering that someone could hack a car
The curse of neural toxicity: AI2 and UW researchers help computers watch their language
UW researchers work to decrease the digital divide in the Puget Sound region
Computer scientists achieve crown jewel of cryptography
Contributing Faculty
- Richard Anderson
- Yejin Choi
- Shyam Gollakota
- Hannaneh Hajishirzi
- Kurtis Heimerl
- Tadayoshi Kohno
- Huijia (Rachel) Lin
- Jamie Morgenstern
- Sewoong Oh
- Shwetak Patel
- Franziska Roesner
- Noah Smith
- Stefano Tessaro
- Yulia Tsvetkov
- Luke Zettlemoyer
- Amy Zhang
Computing for Everyone
Technology can be a great equalizer for people with diverse abilities and experiences — as long as users’ diverse needs and preferences are taken into account. Allen School researchers are leading the way when it comes to the development of new tools and techniques for improving how people interact with technology and the world around them. From making it easier for people to navigate their communities, to ensuring emerging technologies are inclusive of diverse cultures and languages, to assisting those with mobility impairments to complete everyday tasks — and much more! — we produce innovations that will have a tangible, positive impact on people’s lives and extend the benefits of computing to all.
Sample Projects & Papers
- A smartphone app for customized pedestrian trip-planning based on users’ individual mobility profiles
- A smartwatch app that alerts d/Deaf and hard-of-hearing users of sounds occurring in their environment
- Crowdsourced sidewalk accessibility information to improve city planning, build more inclusive mapping tools, and automate the identification of accessibility issues
- A low-cost, portable interaction system for enhancing tactile interactions on mobile phones for users who are blind or low vision
- The introduction of interaction proxies to enable third-party developers to improve mobile app accessibility
- An autonomous robot-assisted feeding system for people with upper-extremity mobility impairments
- Approaches for incorporating backchanneling behavior to enhance interactions with social robots intended for companionship, mental health support, and social-emotional development
- A system for enabling smooth human-robot handovers of arbitrary objects
- An exploration of how web design localization contributes to the user experience across different countries
- An analysis of how cultural values embedded in the design of online communities shape user participation
- The introduction of speaker-grounded neural models of code-switching to advance more adaptable and personalized natural language processing tools