By Matthia SabatelliSource:
[arXiv cs.LG](https://arxiv.org/list/cs.LG/recent)
arXiv:2609.26018v1 Announce Type: new
Abstract: Many learning problems combine a dominant regularity with systematic exceptions. Motivated by U-shaped learning in language acquisition, we study this interaction in controlled quasiregular
regressionproblems where regular and exceptional solutions are explicitly known. Neural networks can partially acquire exceptions, subsequently regress toward the dominant regularity, and finally recover. This overregularization becomes substantially stronger when exceptions are rare, despite their early acquisition, but does not emerge equally across all regularities considered. Our results isolate a simple form of competition between regularities and exceptions during neural learning. Get AI news in your inbox
Daily digest of what matters in AI.