NeurosciencebioRxivPreprint
Representational geometry as a fidelity metric for connectome-constrained networks: evidence from the Drosophila visual system
Michael George Zhou, Jennifer Hasler
The paper introduces representational geometry as a metric to assess biological fidelity in neural networks, showing it distinguishes real from random wiring before training but not after both are trained to task adequacy.
DOI 10.64898/2026.06.10.731214