Applying the Causal Roadmap to Optimal Dynamic Treatment Rules with Lina Montoya - #506
Published:Aug 2, 2021 17:20
•1 min read
•Practical AI
Analysis
This article summarizes a podcast episode from Practical AI featuring Lina Montoya, a postdoctoral researcher. The episode focuses on Montoya's research applying Optimal Dynamic Treatment (ODT) to the US criminal justice system. The discussion covers neglected assumptions in causal inference, the causal roadmap developed at UC Berkeley, and how Montoya uses a "superlearner" algorithm to estimate ODT rules. The article highlights the application of advanced AI techniques to real-world problems and the importance of understanding causal relationships for effective interventions.
Key Takeaways
Reference
“The article doesn't contain a direct quote.”