From Stochastische Texte to t-Distributed Stochastic Neighbor Embedding (t-SNE)

What:
Talk
When:
Tuesday 14 Aug 03:45 PM (1 hour 15 minutes)
Where:
Discussion:
0

Kafka described a foundational labyrinth, -- identity adrift in an implacable circuit -- that Theo Lutz used to seed his "Stochastische Texte" generator. Contemporary techniques, and algorithms such as t-Distributed Stochastic Neighbor Embedding (t-SNE), operate on integrated circuits and offer a topological overview of text corpuses, speckled node- spaces where writing styles and hyperparameters influence proximity. In this talk I’ll examine a few implications of dimensionality reduction (such as t-SNE) for reading. Flattened complexity has continuity with formal critique, topology extends hippocampal mapmaking, hyperparameters detect prosodic patterns, and the implacable blindness of identity encounters algorithm in the labyrinth of language.

Participant
Glia.ca

My Schedule

Add to Your Schedule