clustering
The tendency of productive activity, especially in tech, finance, and creative industries, to concentrate geographically because proximity raises productivity (agglomeration economies).
As defined in Tyler Cowen on Stubborn Attachments, Prosperity, and the Good Society
Other definitions
An unsupervised machine learning method that groups data points by similarity without predefined labels.
Tyler Cowen & Daniel Gross - Identifying Talent - [Invest Like the Best, EP. 277]
Where it comes up · 5
- Tyler Cowen on Stubborn Attachments, Prosperity, and the Good Society3 claims
- OpenAI Whisper Speaker Diarization - Transcription with Speaker Names2 claims
- Building Europe’s Next Software Champion1 claim
- How a Geopolitical Analyst Predicts the Outcome of War | Odd Lots1 claim
- Tyler Cowen & Daniel Gross - Identifying Talent - [Invest Like the Best, EP. 277]1 claim
Most important claims · 8 of 8Search all claims for “clustering” →
- Clustering raises productivity even though you only interact with a few people directly, because there is an aggregation mechanism for information: the best ideas get passed along through chains of people you each speak to, so you are in closer contact with the best ideas relative to the region, and rapid hiring of scarce talent is only feasible in the cluster.Tyler Cowen on Stubborn Attachments, Prosperity, and the Good Society
- Speaker identification in the pipeline uses clustering technique, requiring users to specify the exact number of speakers; incorrectly specifying the speaker count (e.g., saying three speakers when there are only two) will confuse the model and cause incorrect speaker labeling.OpenAI Whisper Speaker Diarization - Transcription with Speaker Names
- Social science has not applied machine learning methods in depth, such as clustering text and images from people's online behavior, and the field remains far from the rigorous tooling used in industry, leaving a gap Gross calls a missing OpenAI for psychology.Tyler Cowen & Daniel Gross - Identifying Talent - [Invest Like the Best, EP. 277]
- Clustering non-base-case scenarios can reveal that the aggregate probability of non-base outcomes is higher than the base case, which is important for clients evaluating tail risks; example: a Uruguay election where the base case was market-constructive but all non-base scenarios aggregated to a higher probability of market-negative outcome.How a Geopolitical Analyst Predicts the Outcome of War | Odd Lots
- The reason clustering matters more now than decades ago is that economic growth has concentrated in clustering-intensive sectors—information technology, finance, creative industries—unlike spread-out activities such as cement production, while increased credentialism and globalization further raise the value of being in major cities.Tyler Cowen on Stubborn Attachments, Prosperity, and the Good Society
- The benefits of clustering smart people together in cities like San Francisco or New York have turned out to be much stronger than most economists expected, especially in information technology, finance, and creative industries, making constant building-restriction policy far more costly because overflow workers in places like Atlanta or Tulsa are far less productive than anticipated.Tyler Cowen on Stubborn Attachments, Prosperity, and the Good Society
- After creating speaker embeddings, the pipeline uses agglomerative clustering (from scikit-learn) to group embeddings by speaker, with the number of clusters determined by the user-specified speaker count.OpenAI Whisper Speaker Diarization - Transcription with Speaker Names
- Chapters does not mandate a public-sector focus but tells platforms to find clusters along value chains, and public-sector clusters have emerged as scalable because success cases are in that ecosystem.Building Europe’s Next Software Champion