There is no regular lab to complete during this week’s lab period. Rather, we will use the lab period to begin work on the mini-project.
Task: Read the mini-project description. Note the deadlines and ask if you have any questions.
Task: Start exploring possible datasets and brainstorming questions you’d like to investigate. Talk to other students to figure out potential pairs (or groups of three). Talk to the instructor and coaches to get feedback on ideas or to get help with a dataset you’re trying out.
And remember that at 4 p.m., there’s a talk, “Biases in Data & Other Threats to Validity of Predictive Models” in Rocky 310. This is the abstract:
Dr Kristina Lerman (USC Information Sciences Institute) will give a talk about how the correlations between the traits, behaviors, and data collection strategies for heterogeneous data can bias data and confound analysis. Using COVID-19 data as an example, she describes a mathematical framework for de-biasing data that addresses these threats to validity of predictive models. The framework promises to learn unbiased models even in analytically challenging datasets.
If you’d like to hear about some interesting research, I encourage you to attend the talk!