Running experiments when there are network effects
Linear Digressions
Traditional A/B tests assume that whether or not one person got a treatment has no effect on the experiment outcome for another person. But thatβs not a safe assumption, especially when there are network effects (like in almost any social context, for instance!) SUTVA, or the stable treatment unit value assumption, is a big phrase for this assumption and violations of SUTVA make for some pretty interesting experiment designs. From news feeds in LinkedIn to disentangling herd immunity from individual immunity in vaccine studies, indirect (i.e. network) effects in experiments can be just as big as, or even bigger than, direct (i.e. individual effects). And this is what we talk about this week on the podcast. Relevant links: http://hanj.cs.illinois.edu/pdf/www15_hgui.pdf https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2600548/pdf/nihms-73860.pdfNext Episodes
Zeroing in on what makes adversarial examples possible @ Linear Digressions
π 2020-01-20 03:41 / β 00:22:51
Unsupervised Dimensionality Reduction: UMAP vs t-SNE @ Linear Digressions
π 2020-01-13 01:53 / β 00:29:34
Data scientists: beware of simple metrics @ Linear Digressions
π 2020-01-05 23:54 / β 00:24:47
Communicating data science, from academia to industry @ Linear Digressions
π 2019-12-30 02:53 / β 00:26:15
Optimizing for the short-term vs. the long-term @ Linear Digressions
π 2019-12-23 03:50 / β 00:19:24