Selective Reconstruction and Emergent Geometry from Complexity Constraints

We propose a modification of holographic tensor network models in which bulk geometry emerges not only from entanglement structure but also from finite physical resource constraints. By assigning a reconstruction cost to bulk operators and introducing a constrained selection principle, we show that the effective geometry corresponds to the subset of operators reconstructible within finite time and energy. This leads to observer-dependent bulk regions and a dynamical notion of geometry governed by reconstruction speed. We demonstrate these ideas in a weighted tensor network model, derive a continuum limit, and show consistency with the linearized Einstein equation.