In our final integrative lab, we brought together different aspects of our hands-on exploration of computational social science. Which task did we NOT perform?

Prof. Glymour from UCSF told us about a case when she used “natural experiments”’ in social epidemiology. She used a double comparison of different groups. Why? And what did she do?

In the lecture, we encountered the concept of an invariant distribution (also known as stationary distribution/measure). What is it and in what context did we encounter it?

A company rolls out an AI co-pilot for strategy writing. Which trade-off should they anticipate and manage, according to a case study with business consultants we’ve reviewed in lecture?

Researcher FindItOut observed specific data showing that social influence extends to networks of 2,000 people. She then created a general theory to explain what she observed in her data. This process of reasoning from specific observations to general theory is an example of:

In the lecture, we reviewed a very short (6min) history of science. How did Warren Weaver characterize the three historically consecutive waves in science’s ability to handle different types of problems?

You are designing a social media platform to maximize the speed at which content spreads. Based on the diffusion simulations from the lecture, which network architecture would you implement?

In the baseline Schelling model shown in lecture, which behavioral rule was NOT assumed (and would change the mechanism from “random relocation” to “active searching”)?

What does the Sugarscape model (developed by Epstein and Axtell in “Growing Artificial Societies”?) primarily illustrate about the concept of social emergence?

According to the so-called godfather of AI, Nobel Prize winner Geoffrey Hinton, what two factors make artificial neural networks fundamentally more scalable than biological human learning?