SyncAI.news, a Varaisys broadcasting
Canonical locks that encode part-whole hierarchies
RM

Rajat Modi, Yogesh Singh Rawat

· 1 min read

ResearcharXiv cs.AI

Canonical locks that encode part-whole hierarchies

arXiv:2609.26046v1 Announce Type: new Abstract: One of the challenges in representational learning is how to encode part-whole hierarchies in a neural net. Prior works rely on flattening tree-like structures into string-like sequences and training a sequence-to-sequence model via autoregression. While such a representation works for parse-trees in NLP, it is not entirely clear how to make it work for images. Thus, we propose a geometric primitive called canonical locks. The key idea is that parts/wholes can be modelled as higher-dimensional vectors ($d \geq 4$), and information can be encoded in their relative phase differences. Inductively, the net consists of positionally-bound bottom-up and top-down neural fields, which drive each other to achieve a state of thermal equilibrium. Additionally, we show the existence of a few symmetrical configurations in the net. The computational iterations taken to break these symmetries depend on the angle between parts/wholes arranged on a disk (or more precisely a ring) in higher dimensions. It also appears to have connections to the psychological phenomenon of mental rotation.

Original source

This story was published by arXiv cs.AI and written by Rajat Modi, Yogesh Singh Rawat. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on arxiv.org

Similar News