• This website continues to be under construction -- if you can't find something here, please also try my old website.

    Interests

    I head the Data to Actionable Knowledge (DtAK) group at Harvard Computer Science.  We use probabilistic methods to address many decision-making scenarios, with a focus on applications related to healthcare.  Our work falls into three major areas:

    • Probabilistic modeling and inference (especially Bayesian models): How can we characterize the uncertainty in large, heterogeneous data?  How can we fit models that will be useful for downstream decision-making?  How can we build models and inference techniques that will behave in expected and desired ways?

    • Decision-making under uncertainty (especially sequential decision-making): How can we optimize policies given batches of heterogeneous data?  How can we provide useful information, even if we can't solve for a policy?  How can we characterize the limits of our ability to provide decision support?

    • Interpretability and Statistical methods for validation: How can we estimate the quality of a policy from batch data?  How can we expose key elements of a model or policy for expert inspection? 

    Short Bio

    Finale Doshi-Velez is a John L. Loeb associate professor in Computer Science at the Harvard Paulson School of Engineering and Applied Sciences.  She completed her MSc from the University of Cambridge as a Marshall Scholar, her PhD from MIT, and her postdoc at Harvard Medical School.  Her interests lie at the intersection of machine learning, healthcare, and interpretablity.
     
    Selected Additional Shinies: AFOSR YIP and NSF CAREER recipient; Sloan Fellow; IEEE AI Top 10 to Watch 

    Getting in touch

    In person: My office hours are generally 3:30-4:30 on Wednesdays; 3-4:30 in the spring term while CS181 is in session.  When I have my office hour, please come in (don't wait outside!!) even if other people are inside.  I will ask everyone about what they want to discuss, and discuss similar topics together.  I usually give class-related questions first priority, then switch over to general topics and then back again.  If you have a form that needs signing, just interrupt.  I will continue holding office hours via zoom for the rest of the semester.  Link: https://harvard.zoom.us/j/768366874 .  In general, I will prioritize CS181 and undergrad research.  Like most of you, I'm still trying to figure out new routines; if the times need to change I will post here.

    Email: I get a lot of email, and my inbox is generally an absolute disaster.  Unfortunately, many times I don't even have a chance to read all my email.  Please help me by reading my website first, having an informative subject line, and pinging again if you feel like I might have lost your mail.  Office hours don't require an email first, and are a great time to also catch me in person.

    Phone: I don't have an office phone any more.  Please email or stop by in person.

    Scheduling a qual, mini-qual, PhD defense, Master's defense, etc.

    I get a lot of requests, and I am both eager to be part of this important milestone and also going crazy with scheduling.  Thus, I am now offering to be part your qual, mini-qual, PhD defense, Master's defense, or similar type of event at three sets of dates each year, roughly fall, winter, and spring. 
     
    For 2018-2019, these will be: September 17 and 18; January 21 and 22, and May 20 and 21.  The Monday slots will be 11-1, 1-3, and 3-4:30; the Tuesday slots will be 8-10, 10-12, 12-2, and 2-4.  [January 21 3-4:30 , January 22 8-10, 10-12 , May 21 10-12, 12-2, 2-4 May 20 11-1, 1-3, 3-4:30 is taken.]
     
    For 2019-2020, I will use the same time slots as before.  The dates will be: September 9 and 10 [Sep. 9 11-1 is taken. Sept. 10 2-4pm is taken.].  January 27 and 28 [January 27 11-1pm, 1-3pm, 3pm-4:30pm* is taken, January 28 3-4:30 is taken].  May 4 and 5 [May 4 9:30-11, 11-1, 1-3,* 3-4:30 is taken; May 5 11-1, 1-3, 3-4:30 is taken]