Deep neural networks have surpassed human-level performance for many cognitive tasks. However, it is difficult to understand how the neural networks makes decisions. This research position is on explainable AI, which aims to make neural networks more interpretable to humans. The will involve training your own neural network, testing different attribution algorithms, and designing your own algorithms.
The project is 10 per week. I meet with my undergraduate researchers once per week. The project is one semester but it can potentially be extended to multiple semesters.
Project Dates
Start Date: 1/9/2023 - End Date: 4/28/2023
Students Needed
Type of Project
Individual or team
Student Responsibilities
You are required to work 10 hours per week and attend weekly group meeting.
Time Commitment
10 hour(s)
Student Requirements
There are no specific requirements. Basic programming knowledge would be helpful. Junior and senior students are preferred.
Interested in Working With the Following Programs
For EXCEL URE Students Only
Additional Notes