Deep Neural Regression Collapse

📰 ArXiv cs.AI

Neural Regression Collapse occurs across layers in deep neural regression models, not just the last layer

advanced Published 26 Mar 2026

Action Steps

Identify the phenomenon of Neural Regression Collapse in deep neural regression models
Analyze the occurrence of NRC across different layers and models
Apply insights from NRC to improve model interpretability and performance

Who Needs to Know This

Machine learning researchers and engineers benefit from understanding NRC to improve model interpretability and performance, while data scientists can apply these insights to develop more accurate regression models

Key Insight

💡 Neural Regression Collapse is not limited to the last layer, but occurs across layers in deep neural regression models

Key Takeaways

Neural Regression Collapse occurs across layers in deep neural regression models, not just the last layer

Full Article

Title: Deep Neural Regression Collapse

Abstract:
arXiv:2603.23805v1 Announce Type: cross Abstract: Neural Collapse is a phenomenon that helps identify sparse and low rank structures in deep classifiers. Recent work has extended the definition of neural collapse to regression problems, albeit only measuring the phenomenon at the last layer. In this paper, we establish that Neural Regression Collapse (NRC) also occurs below the last layer across different types of models. We show that in the collapsed layers of neural regression models, features

Read full paper → ← Back to Reads