Mary Gates Research Scholar, Winter 2021
Research Project: Using Self-Supervised Learning in Decoding Electrocorticography
Project Description: Decoding the electrocorticography (ECoG) data from a patient’s brain can be done by labeling specific behaviors in the patient’s data. However, it is very inefficient to manually annotate thousands of hours worth of data, so we are exploring self-supervised learning algorithms to autogenerate behavior labels on the ECoG brain data. The end product is a powerful way of unpacking the unique structures and processes happening inside of our brains!
What have you learned throughout your research project?
Research in general is already very fast-paced. The intersection of data and brain sciences is especially difficult to keep up with in this sense, so a huge part of being a researcher is constantly being up to date with the newest findings and technologies. Even if you have years of schooling, everyone has to be ready to continually learn, learn, learn.
What piece of advice do you have for future applicants?
Don’t let the vastness and depth of a field intimidate you, the quickest way to learn is to dip your toes in the water and try and fail! If we spend too long assessing our own qualifications then we can never put ourselves out there. You have to embrace the possibility of being wrong and failing in order to celebrate the process of discovery!