{"id":"semantic-neighborhoods","title":"Meaningful neighborhoods and the unclassified remainder","regions":["representation"],"question":"What counts as a meaningful region of a representation space, and what remains outside the regions a system can recognize?","formulation":"Face, person, and expression are proposed as overlapping meaning regions. Narrower categories can increase discrimination, while the status of noise or meaningless dimensions remains unsettled.","representations":[{"name":"Overlapping sets of recognized cases","affords":"Expresses intersecting classifications without assuming smoothness.","loses":"Does not describe local geometric operations.","epistemic_kind":"reconstruction"},{"name":"Local geometric charts","affords":"Supports neighborhoods, directions, and changes in resolution.","loses":"Smoothness, dimension, and distance must be established for the chosen data.","epistemic_kind":"reconstruction"}],"tension":"An interpretable region need not be a mathematical manifold; a classifier covering the space need not make everything meaningful.","challenge":"How could a system encounter a case for which its existing meaning regions are inadequate?","development":"active","epistemic_kind":"reconstruction","evidence":["basis-semantic-neighborhoods"],"next_move":{"text":"Define a rejection or unclassified region and compare it with forcing every example into an existing category.","epistemic_kind":"model_proposed"},"open_questions":["What counts as a meaningful region of a representation space, and what remains outside the regions a system can recognize?","How could a system encounter a case for which its existing meaning regions are inadequate?"],"formal_status":"open_problem","version":1,"content_hash":"31e9087dea27c11e77c472d30d5efaa1f535144491da7bcb2fd698b3d718c1bd"}