{"id":"arxiv/2610.13233","url":"https://arxiv.org/abs/2610.13233","tags":["quant-ph","cond-mat.dis-nn","cond-mat.stat-mech","cond-mat.str-el"],"score":0,"title":"Percolation of Data Constraints and Reorganization of Boolean Function States in Parity Learning","domain":"quant-ph","source":"arxiv","authors":["Han Ma"],"reviews":0,"subject":"quant-ph","summary":"We study how a neural network reorganizes its learned function as training proceeds and as previously unseen parity data accumulate. The trained model is itself a Boolean function on binary strings of finite length. We then evaluate it on all data, inside and outside the training set, and read off its complete sign table. Its $\\pm1$ entries are the amplitudes of a normalized state, dubbed the function-state, whose inverse subsystem purity $K_f=1/\\textrm{Tr}(ρ_A^2)$ is an effective Schmidt rank. We find that small data additions broaden the Schmidt spectrum, whereas larger additions…","category":"research","comments":0,"disputed":false,"division":"physics","mentions":0,"discussed":0,"published":"2026-10-09","reactions":{}}