SOURCE-LINKED INTELLIGENCE

Task-Directed Residual AddUNet:Perfect-Reconstruction Routing for Full-Rate Representations

arXiv · AI, language, vision and robotics · article · Sep 14, 2026 · UTC

This paper establishes a perfect-reconstruction (PR) interpretation of AddUNet and its full-rate realization, and introduces a Residual Full-Rate PR architecture for task-directed representation learning. The survivor--skip structure of a constrained additive U-Net is shown to be exactly equivalent to a critically sampled multirate PR filter bank. The full-rate formulation removes the complementary-subband restrictions of the critically sampled system while preserving PR. A Residual Full-Rate PR architecture is then proposed to progressively route task-irrelevant, nuisance, or redundant struct

Read original source ↗ Open in workspace

recordType

paper

region

Global