Skip to yearly menu bar Skip to main content


Poster

Position: The Platonic Representation Hypothesis

Minyoung Huh · Brian Cheung · Tongzhou Wang · Phillip Isola

Hall C 4-9 #308
[ ] [ Project Page ] [ Paper PDF ]
Tue 23 Jul 4:30 a.m. PDT — 6 a.m. PDT
 
Oral presentation: Oral 2A Representation Learning 1
Tue 23 Jul 7:30 a.m. PDT — 8:30 a.m. PDT

Abstract:

We argue that representations in AI models, particularly deep networks, are converging. First, we survey many examples of convergence in the literature: over time and across multiple domains, the ways by which different neural networks represent data are becoming more aligned. Next, we demonstrate convergence across data modalities: as vision models and language models get larger, they measure distance between datapoints in a more and more alike way. We hypothesize that this convergence is driving toward a shared statistical model of reality, akin to Plato's concept of an ideal reality. We term such a representation the platonic representation and discuss several possible selective pressures toward it. Finally, we discuss the implications of these trends, their limitations, and counterexamples to our analysis.

Chat is not available.