Episode 2: Fooling Yourself Less: The Art of Statistical Thinking in AI

In this episode, we hear from Andrew Gelman, a leading statistician and political scientist, who dives into the practical side of data science. He highlights the importance of data quality, simulations, and coding skills, advocating for these aspects over traditional statistical theory. Gelman uses compelling examples to show how simulations can deepen our understanding of statistical concepts. He also tackles the limitations of polling and underscores the need for causal inference in real-world applications. Ultimately, the discussion promotes a hands-on, coding-focused approach to data science, encouraging practitioners to emphasize clarity and rigor in their analyses.