:focal(3277x360:3278x361))
Data science is redefining what’s possible across all fields of knowledge—from astrophysics to history, and everything in between. At Stanford, investments in data science are turning new possibilities into profound discoveries.
:focal(890x1292:891x1293)){kind=spacer}
Stanford Data Science is on a mission to weave data science into the core fabric of Stanford’s research and teaching enterprises.
:focal(1106x972:1107x973))
)
Photo: Evgeni Tcherkasski, Unsplash
Where will the greatest discoveries of the 21st century be made? Like earlier generations of scholars, today’s researchers still peer into telescopes, excavate prehistoric ruins, and observe animals in the wild in their quest for new knowledge. What distinguishes scholars today is their sophisticated use of digital technologies to collect and analyze vast quantities of data that did not exist even a short time ago.
What is dark matter? How can we develop lifesaving drugs faster? Can we better predict and respond to extreme weather?
The rise of massive datasets and advanced computational technologies are not only propelling discovery in the fundamental mysteries of the universe but also enabling researchers to embark on entirely new quests.
Data science offers researchers the tools and techniques to navigate this new digital frontier. In an era marked by a super-abundance of information and rapid advances in computing and artificial intelligence, data science has begun to reshape entire industries and fields. Some of the most exciting work at Stanford is happening at the intersection of data science and a wide array of other disciplines. While astrophysicists work to unravel the universe’s secrets using data from state-of-the-art telescopes, historians reconstruct our past from troves of digitized records. In both cases, data science enables scholars to turn data into insights that change how we think about the world and our place in it.
By bringing talented data scientists to campus, building interdisciplinary communities to spur new lines of research, and offering access to the most advanced computing resources, Stanford Data Science is creating the ecosystem to ensure Stanford remains at the forefront of discovery for years to come—helping to create a better future for all.
Stanford Data Science recruits and nurtures leaders in data science who transcend traditional disciplinary boundaries. These scholars advance the field of data science and pursue innovative applications to drive discovery across campus.
Stanford Data Science connects diverse faculty, staff, students, and partners beyond the university to exchange knowledge and spark new ways of thinking in key priority areas.
Shared computational resources staffed by research data scientists accelerate data-intensive discovery across campus and open the door to breakthroughs that were previously unachievable.
Stanford Data Science recruits and nurtures leaders in data science who transcend traditional disciplinary boundaries. These scholars advance the field of data science and pursue innovative applications to drive discovery across campus.
The Data Science Scholars program supports a community of graduate students from all seven Stanford schools who innovate and apply data science methods in their disciplines. Typically, around 125 students apply for only 13 positions. Alumni of the program have gone on to take positions in academia and industry, bringing data science approaches to areas like the life sciences and environmental protection.
)
Data Science Postdoctoral Fellows are recent doctoral graduates with dual expertise in data science and another discipline. During a one- to two-year program, fellows work with faculty mentors to build new data science methods that advance research in their fields. Each year, interest in the program far exceeds capacity, and fellows have contributed to breakthroughs across the university.
)
Stanford Data Science hires faculty jointly with Stanford schools, departments, and institutes, placing talented data scientists in an ecosystem purpose-built to accelerate discovery. To date, faculty have been appointed to the schools of Engineering and Humanities and Sciences.
)
)
Meet some of the faces of Stanford Data Science.
)
)
)
)
Stanford Data Science connects diverse faculty, staff, students, and partners beyond the university to exchange knowledge and spark new ways of thinking in key priority areas.
Stanford Data Science’s five faculty-led research centers build interdisciplinary communities of practice around themes of common interest. Center affiliates collaborate on research projects and share their latest work in center-led seminars, conferences, and other events. The centers’ numerous external collaborators—including a wide range of tech and consumer product companies and nonprofits—make them important platforms for fostering engagement beyond campus.
)
The Stanford Causal Science Center (SC²) fosters interdisciplinary and industry collaboration around causality and causal inference through seminars and conferences. It also encourages graduate students and postdoctoral researchers to explore and apply causal inference methods across fields including economics, public health, business, law, and education, among others in which the rigorous study of cause and effect is essential to producing useful knowledge.
Faculty leaders: Guido Imbens (Graduate School of Business, Economics), Stefan Wager (Graduate School of Business), and Ramesh Johari (Management Science and Engineering)
CORES promotes open and reproducible data science. Open science (also called “open scholarship” or “open research”) is the practice of widely sharing data and research across disciplines and institutions in a timely manner. It deepens trust in scientific research and fuels collaboration and innovation.
Faculty leader: Russ Poldrack (Chair, Psychology)
SuDS catalyzes the application of data science in sustainability research, providing a forum for faculty and students to share and learn about data-intensive methods. It also offers undergraduates a meaningful capstone experience that involves applying data science skills to real-world research.
Faculty leader: David Lobell (Earth System Science)
Data4Health convenes a university-wide community of researchers and students who use data science to extract biomedical and health care insights. It organizes seminars and other activities to promote data-intensive methods in health care and biomedical research.
Faculty leader: James Zou (Biomedical Data Science)
)
CDU brings together researchers from diverse disciplines to advance our understanding of the universe. Together, CDU participants develop cutting-edge methodologies harnessing data science and AI to extract insights from complex data on the cosmos. The center aims to rapidly demonstrate the relevance of those innovations for diverse scientific domains.
Faculty leaders: Risa Wechsler (Physics/Astrophysics) and Susan Clark (Astrophysics).
)
)
The two wings of the Computing and Data Science (CoDa) building – one a rectangular shape, one an oval – connote the 1s and 0s binary code theme. Photo: Andrew Brodhead
Shared computational resources staffed by research data scientists accelerate data-intensive discovery across campus and open the door to breakthroughs that were previously unachievable.
Powerful computers have become as foundational to scholarship in the 21st century as libraries have been for millennia. Research in a wide array of fields requires immense computational power to process large datasets and run sophisticated simulations. Stanford’s new high-performance GPU cluster, Marlowe, is available to investigators across the university.
Named after the fictional detective with a knack for solving mysteries, Marlowe is staffed by highly skilled research data scientists who collaborate with research teams on how to use the platform to advance their work. Since the system enables more sophisticated machine-learning models and simulations than are feasible with Stanford’s other computing resources, it has the potential to dramatically accelerate research and lead to breakthroughs we have yet to imagine.
Below are just a few of Marlowe’s exciting applications based on initial testing:
)
Marlowe enables astronomers to run complex simulations and process detailed images while studying phenomena like dark matter and galaxy evolution, shedding light on the structure and dynamics of the universe.
In the early stages of drug development, investigators search massive chemical libraries for useful molecular structures. Marlowe accelerates and reduces the cost of this virtual screening process, enabling quicker identification of potentially lifesaving drugs.
Predicting the effects of climate change and extreme weather events involves stimulating atmospheric processes and developing high-resolution climate and weather models. Marlowe enables researchers to build more accurate models in less time.
:focal(1744x584:1745x585))
What if we were fluent in the language of biology? Brian Hie is on a mission to translate life’s code, predict viral evolution, and revolutionize medicine.
:focal(1071x1527:1072x1528)){kind=spacer}
Guido Imbens uses data science to search for cause and effect in the wild.
)
)
What future insights and innovations will be possible because of our investments in data science today? Stanford is building a data science ecosystem for the 21st century, where data-intensive research, application, and education drive new discoveries. We hope you will join us to ensure Stanford continues to redefine what is possible and shapes the world for the better.
)
)
The scope of Stanford’s vision matches the scope of the challenges we face. Because Stanford has built excellence in an unusually broad range of fields, and because we’ve connected those fields.