Preventing Model Collapse Under Overparametrization: Optimal Mixing Ratios for Interpolation Learning and Ridge Regression

arXiv:2509.22341v2 Announce Type: replace-cross Abstract: Model collapse occurs when generative models degrade after repeatedly training on their own synthetic outputs. We study this effect in overparameterized linear regression in a setting where each iteration mixes fresh re...

🔗 Read more: https://arxiv.org/abs/2509.22341

#News #Policy #Biology #Academic