English
Related papers

Related papers: Preventing Model Collapse Under Overparametrizatio…

200 papers

In this work, we investigate the behavior of ridge regression in an overparameterized binary classification task. We assume examples are drawn from (anisotropic) class-conditional cluster distributions with opposing means and we allow for…

Machine Learning · Statistics 2025-03-12 Alexander Tsigler , Luiz F. O. Chamon , Spencer Frei , Peter L. Bartlett

Learning discrete distributions from i.i.d. samples is a well-understood problem. However, advances in generative machine learning prompt an interesting new, non-i.i.d. setting: after receiving a certain number of samples, an estimated…

Information Theory · Computer Science 2026-01-06 Millen Kanabar , Michael Gastpar

Training deep neural networks for classification often includes minimizing the training loss beyond the zero training error point. In this phase of training, a "neural collapse" behavior has been observed: the variability of features…

Machine Learning · Computer Science 2023-05-30 Tom Tirer , Haoxiang Huang , Jonathan Niles-Weed

What happens when generative machine learning models are pretrained on web-scale datasets containing data generated by earlier models? Some prior work warns of "model collapse" as the web is overwhelmed by synthetic data; other work…

Machine Learning · Computer Science 2025-03-19 Joshua Kazdan , Rylan Schaeffer , Apratim Dey , Matthias Gerstgrasser , Rafael Rafailov , David L. Donoho , Sanmi Koyejo

Deep metric learning (DML) aims to learn a discriminative high-dimensional embedding space for downstream tasks like classification, clustering, and retrieval. Prior literature predominantly focuses on pair-based and proxy-based methods to…

Computer Vision and Pattern Recognition · Computer Science 2024-07-04 Xiruo Jiang , Yazhou Yao , Xili Dai , Fumin Shen , Xian-Sheng Hua , Heng-Tao Shen

A conventional wisdom in statistical learning is that large models require strong regularization to prevent overfitting. Here we show that this rule can be violated by linear regression in the underdetermined $n\ll p$ situation under…

Statistics Theory · Mathematics 2024-06-06 Dmitry Kobak , Jonathan Lomond , Benoit Sanchez

Prototypical self-supervised learning methods consistently suffer from partial prototype collapse, where multiple prototypes converge to nearly identical representations. This undermines their central purpose -- providing diverse and…

Random matrix theory has become a widely useful tool in high-dimensional statistics and theoretical machine learning. However, random matrix theory is largely focused on the proportional asymptotics in which the number of columns grows…

Statistics Theory · Mathematics 2025-06-23 Chen Cheng , Andrea Montanari

From benign overfitting in overparameterized models to rich power-law scalings in performance, simple ridge regression displays surprising behaviors sometimes thought to be limited to deep neural networks. This balance of phenomenological…

Machine Learning · Statistics 2026-05-08 Alexander Atanasov , Jacob A. Zavatone-Veth , Cengiz Pehlevan

Researchers in empirical machine learning recently spotlighted their fears of so-called Model Collapse. They imagined a discard workflow, where an initial generative model is trained with real data, after which the real data are discarded,…

Machine Learning · Computer Science 2024-10-31 Apratim Dey , David Donoho

Generative artificial intelligence is rapidly transforming the supply side of training data: an increasing share of new tokens, images, and structured records is produced by previous-generation models rather than by human originators.…

General Economics · Economics 2026-05-21 Gustav Olaf Yunus Laitinen-Fredriksson Lundström-Imanov

Classical statistical learning theory predicts a U-shaped relationship between test loss and model capacity, driven by the bias-variance trade-off. Recent advances in modern machine learning have revealed a more complex pattern,…

Machine Learning · Statistics 2026-04-23 Yin Liu , Jianwen Cai , Didong Li

High-quality data is essential for training large generative models, yet the vast reservoir of real data available online has become nearly depleted. Consequently, models increasingly generate their own data for further training, forming…

Machine Learning · Computer Science 2025-02-27 Shi Fu , Yingjie Wang , Yuzhu Chen , Xinmei Tian , Dacheng Tao

We analyze the prediction error of ridge regression in an asymptotic regime where the sample size and dimension go to infinity at a proportional rate. In particular, we consider the role played by the structure of the true regression…

Statistics Theory · Mathematics 2021-03-09 Dominic Richards , Jaouad Mourtada , Lorenzo Rosasco

Traditionally in regression one minimizes the number of fitting parameters or uses smoothing/regularization to trade training (TE) and generalization error (GE). Driving TE to zero by increasing fitting degrees of freedom (dof) is expected…

Machine Learning · Computer Science 2019-06-11 Partha P Mitra

In recent years, there has been a significant growth in research focusing on minimum $\ell_2$ norm (ridgeless) interpolation least squares estimators. However, the majority of these analyses have been limited to an unrealistic regression…

Statistics Theory · Mathematics 2024-06-14 Sungyoon Lee , Sokbae Lee

We consider correlated \emph{factor} regression models (FRM) and analyze the performance of classical ridge interpolators. Utilizing powerful \emph{Random Duality Theory} (RDT) mathematical engine, we obtain \emph{precise} closed form…

Machine Learning · Statistics 2024-06-14 Mihailo Stojnic

The inference performance of the pseudolikelihood method is discussed in the framework of the inverse Ising problem when the $\ell_2$-regularized (ridge) linear regression is adopted. This setup is introduced for theoretically investigating…

Disordered Systems and Neural Networks · Physics 2021-10-19 Xiangming Meng , Tomoyuki Obuchi , Yoshiyuki Kabashima

Self-distillation (SD) is the process of retraining a student on a mixture of ground-truth labels and the teacher's own predictions using the same architecture and training data. Although SD has been empirically shown to often improve…

Statistics Theory · Mathematics 2026-02-20 Hien Dang , Pratik Patil , Alessandro Rinaldo

What scaling limits govern neural network training dynamics when model size and training time grow in tandem? We show that despite the complex interactions between architecture, training algorithms, and data, compute-optimally trained…

Machine Learning · Computer Science 2025-07-08 Shikai Qiu , Lechao Xiao , Andrew Gordon Wilson , Jeffrey Pennington , Atish Agarwala