Does the Brain Infer Invariance Transformations from Graph Symmetries?

TL;DR

This paper explores how the brain might learn invariance to perceptual changes through graph symmetries in neural connections, supported by natural data correlations and consistent neural architecture predictions.

Contribution

It proposes a biologically plausible model where the brain infers invariance transformations from graph symmetries in neural connectivity.

Findings

Supports the hypothesis with natural data correlations

Predicts neural connectivity architecture aligned with empirical observations

Suggests a mechanism for invariance learning in sensory cortex

Abstract

The invariance of natural objects under perceptual changes is possibly encoded in the brain by symmetries in the graph of synaptic connections. The graph can be established via unsupervised learning in a biologically plausible process across different perceptual modalities. This hypothetical encoding scheme is supported by the correlation structure of naturalistic audio and image data and it predicts a neural connectivity architecture which is consistent with many empirical observations about primary sensory cortex.