{"id":"arxiv/2607.07206","url":"https://arxiv.org/abs/2607.07206","tags":["cs.LG","math.OC"],"score":0,"title":"Causal Optimizer Interaction Calculus: Hidden Geometric Relaxation and Identifiable Interventions","domain":"cs.LG","source":"arxiv","authors":["Zavier Li"],"reviews":0,"subject":"cs.LG","summary":"Optimizer experiments observe responses to algorithmic configurations without uniquely revealing hidden mechanisms. We develop a causal optimizer interaction calculus that separates pathwise realization, Mobius decomposition, and experimental identification. Under a fixed innovation coupling, every finite-horizon innovation-driven optimizer admits a behaviorally minimal pathwise realization. For any finite effect support and intervention design, an incidence operator gives the complete observational gauge, exact identifiability, sharp quotient stability, held-out predictions, and exact…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-07-08","reactions":{}}