An essential capability of human intelligence is being able to learn “how to learn” (meta-learn) and thus being effective at transferring to novel tasks (with few data examples) by leveraging prior experience. Meta-learning is a new rich field to solve this domain, however, in light of recent evidence it is unclear when and where meta-learning methods should work. The task for this project will be to explore this and eventually use this knowledge to improve the progress in the field of meta-learning.
Compensation: Opportunity available for either credit or pay (must be negotiated)
If interested provide **all options below**:
- [required] Send Brando an emailing briefly explain why you are interested and how you can contribute to the project and the confirmation that a previous mentor/faculty in a medium/long-term project will send me a reference letter. Also, communicate to me the number of hours per week you can commit to the project.
- [required] Send Brando your CV
- [required] Send Brando your transcript.
- [required] Letter of recommendation with emphasis on your ability to work independently on medium/long term (>= 3 months) projects (e.g. other research experience, internship project) or potential in engaging in undergraduate research. Make sure your letter writer (and not you) send the letter directly to me to miranda9@illinois.edu. Note: letter of reference is not optional.
Contact: Brando Miranda (Faculty Advisor: Sanmi Koyejo)
Email: miranda9@illinois.edu Website: https://brando90.github.io/brandomiranda
