CSCI 461: AI In Sustainable Development
Taught by Bistra Dilkina, the Co-Director of the Center for AI in Society, CSCI 461 is a course that is only offered for the fall semester. In our course, Dr. Dilkina runs through many of the key machine learning algorithms while showing how computer scientists collaborate with domain experts to have a tangible impact. We’ve covered how researchers have utilized machine learning to predict climate-related sea level rise, optimize sample locations in Alaska to measure permafrost melt, catch illegal animal poachers, detect contaminations in public water systems, and more! Yet what makes this class interesting is that we learn about this work so we can apply it in our own research!
We designed our project around the National Youth in Transition Dataset (NYTD). Essentially, researchers tracked youth outcomes from the foster care systems across the country. Researchers surveyed employment, education status, incarceration rates, drug use, and housing status, as well as independent living services used to facilitate the transition out of the foster care system. In our project, we will utilize machine learning to produce fair and equitable models that predict the impact of service utilization on adult outcomes to identify where policymakers can allocate resources on the state level.
Originally, we wanted to collaborate with a professor at Keck School of Medicine to identify how mindfulness interventions can help reduce nicotine dependence. However, unfortunately, complications in the study made it so the collaboration fell through. Ultimately, our team could pivot to analyze NYTD, which was a blessing in disguise. As a team, we are more interested in active policy-making with tangible impact. As we gather more results, I’ll keep everyone updated!
I’m thrilled that I’m able to pursue this line of work as an undergraduate and I’m looking forward to sharing the projects I’ve been able to contribute to at the end of the semester 🙂