ZARZA We are Zarza, the prestigious firm behind major projects in information technology.
Listen "Episode 14.20: Polysemantic Encoding"
Episode Synopsis
When nodes are involved in multiple encodings, the brain or networks can encode far more concepts but at the expense of being harder or impossible to interpret. Accordingly, there is a trade-off between power and interpretability, even between power and security: how do we know what polysemic neural nets are doing or “thinking”?
More episodes of the podcast Unmaking Sense
Episode 14.33: Language and the Self
30/07/2025
Episode 14.32: Inverse Hypostatisation?
28/07/2025
In God we trust