DRI-ICE Research Group: Deploying Responsible Intelligence In Changing Environments
Haoran Zhang (PhD)
Taylor Killian (PhD)
Kumail Alhamoud (PhD)
Anurag Vaidya (PhD)
Alice Jin (PhD)
Nathan Ng (PhD)
Yuxin Xiao (PhD)
At U. of Toronto
Sujay Nagaraj (PhD)
Sana Tonekaboni (PhD)
Prathyush Parvatharaju (PhD)
Ramesh Doddaiah (PhD)
Dongliang Guo (PhD)
Menxuan Hu (PhD)
Zihan Guan (PhD)
Bryan Christ (PhD)
Kyle O'Brien (MSFT)
While our group studies impactful, fast-paced, and exciting problems, we emphasize healthy work-life balance and general wellbeing to maintain sustainable quality of life.
I aim to broaden participation in machine learning research in all aspects: folks from all walks of life are welcome and I do my best to accommodate everyone's situation.
Successful research should be fun to conduct and involve diving deep into details. To facilitate this, we work together to define concrete problems we are both excited about.
I encourage and seek independence at all stages of your degree.
Supervision style: I meet each student 1:1 for 1 hour per week. This meeting is student-led and often focuses on research updates and feedback on research. I also try to be available as much as possible for paper-writing and other support.
A big component of doing a PhD is self-discovery, so I encourage students to try new things and read broadly. One way I facilitate this is through a weekly paper-reading group.
Joining the group
We are recruiting two Data Science Ph.D. students to start Fall 2024
I am recruiting Ph.D. students to join my group in the University of Virginia's School of Data Science to start in Fall 2024 and work on responsible machine learning in ever-changing environments. I'm especially looking for students who want to work on:
Post-training interventions for pre-trained models. Example: How do we make targeted edits to big models without incurring costly retraining?
Pre-training multi-modal models. Example: How can we leverage powerful language representations to bolster low-resource data?
Adapting models to ever-changing environments. Example: How can we train models that are robust to distribution shifts?
Who is an ideal candidate? I'm seeking students who
Are enthusiastic to perform high-impact research creatively and with an eye for detail.
Have prior experience conducting research and communicating your findings, ideally through a published paper.
Have prior experience in math (linear algebra/probability/optimization) and/or ML programming (pytorch/numpy/huggingface/pandas etc).
You do not need to meet any or all criteria to apply!
Students will receive frequent one-on-one mentorship and research experience.
I will actively invest in your intellectual, professional, and personal growth by supporting conference travel, helping you network, and providing useful resources.
I prioritize mental health by encouraging students to foster work-life balance, take breaks, and set professional communication boundaries.
Please apply directly to the Data Science PhD program.
If you think you are an excellent fit, please send me an email at hartvigsen[at]virginia.edu including your CV/resume and undergraduate transcript. But I cannot guarantee individual responses due to high request volume.
Current UVA or MIT students
If you are a current student at UVA or MIT looking to do a project with me, please send me an email with "[CURRENT STUDENT]" in the subject line.