SOURCE-LINKED INTELLIGENCE

Functional Degeneracy in Neural Networks: Measurement and Pruning

arXiv · AI, language, vision and robotics · article · Aug 31, 2026 · UTC

A central question in modern machine learning is how much a trained model can be compressed without changing its behavior, to reduce the memory, compute and energy required to deploy it. To study this, we quantify functional degeneracy through the behavioral recovery rank, defined as the number of leading behavioral-Hessian eigendirections required to recover a trained model's performance. Using the behavioral recovery rank as a geometric benchmark for compression, we find that structural and magnitude pruning retain more degrees of freedom, even after the task is saturated. This gap suggests

Read original source ↗ Open in workspace

recordType

paper

region

Global