Image de couverture de livre

Risk Phase Transitions in Spiked Regression: Alignment Driven Benign and Catastrophic Overfitting

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Li, Jiping, Sonthalia, Rishi
Format:	Preprint
Publié:	2025
Sujets:	Machine Learning Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Generalization for Least Squares Regression With Simple Spiked Covariances
par: Li, Jiping, et autres
Publié: (2024)

Benign, Tempered, or Catastrophic: A Taxonomy of Overfitting
par: Mallinar, Neil, et autres
Publié: (2022)

Benign Overfitting for Regression with Trained Two-Layer ReLU Networks
par: Park, Junhyung, et autres
Publié: (2024)

Low Rank Gradients and Where to Find Them
par: Sonthalia, Rishi, et autres
Publié: (2025)

Benign Overfitting in Adversarial Training for Vision Transformers
par: Zhang, Jiaming, et autres
Publié: (2026)

A Classical View on Benign Overfitting: The Role of Sample Size
par: Park, Junhyung, et autres
Publié: (2025)

Double Descent and Overfitting under Noisy Inputs and Distribution Shift for Linear Denoisers
par: Kausik, Chinmaya, et autres
Publié: (2023)

Least Squares Regression Can Exhibit Under-Parameterized Double Descent
par: Li, Xinyue, et autres
Publié: (2023)

Unveiling the Backdoor Mechanism Hidden Behind Catastrophic Overfitting in Fast Adversarial Training
par: Zhao, Mengnan, et autres
Publié: (2026)

Catastrophic Overfitting, Entropy Gap and Participation Ratio: A Noiseless $l^p$ Norm Solution for Fast Adversarial Training
par: Mehouachi, Fares B., et autres
Publié: (2025)

The Viscosity of Logic: Phase Transitions and Hysteresis in DPO Alignment
par: Pollanen, Marco
Publié: (2026)

Control of Overfitting with Physics
par: Kozyrev, Sergei V., et autres
Publié: (2024)

Catastrophic Forgetting Mitigation Through Plateau Phase Activity Profiling
par: Mashiach, Idan, et autres
Publié: (2025)

Do Deep Neural Network Solutions Form a Star Domain?
par: Sonthalia, Ankit, et autres
Publié: (2024)

On Regularization via Early Stopping for Least Squares Regression
par: Sonthalia, Rishi, et autres
Publié: (2024)

GRC-Net: Gram Residual Co-attention Net for epilepsy prediction
par: You, Bihao, et autres
Publié: (2025)

How to Mitigate Overfitting in Weak-to-strong Generalization?
par: Shi, Junhao, et autres
Publié: (2025)

Towards Mitigating Architecture Overfitting on Distilled Datasets
par: Zhong, Xuyang, et autres
Publié: (2023)

Residual Stream Analysis of Overfitting And Structural Disruptions
par: Liu, Quan, et autres
Publié: (2026)

Dimensional Characterization and Pathway Modeling for Catastrophic AI Risks
par: Chin, Ze Shen
Publié: (2025)

How Catastrophic is Your LLM? Certifying Risk in Conversation
par: Wang, Chengxiao, et autres
Publié: (2025)

Representations Shape Weak-to-Strong Generalization: Theoretical Insights and Empirical Predictions
par: Xue, Yihao, et autres
Publié: (2025)

Consequentialist Objectives and Catastrophe
par: Marklund, Henrik, et autres
Publié: (2026)

Have Large Language Models Learned to Reason? A Characterization via 3-SAT Phase Transition
par: Hazra, Rishi, et autres
Publié: (2025)

Benign Overfitting and the Geometry of the Ridge Regression Solution in Binary Classification
par: Tsigler, Alexander, et autres
Publié: (2025)

Superintelligent Agents Pose Catastrophic Risks: Can Scientist AI Offer a Safer Path?
par: Bengio, Yoshua, et autres
Publié: (2025)

Friend or Foe? Harnessing Controllable Overfitting for Anomaly Detection
par: Qian, Long, et autres
Publié: (2024)

On the Implicit Reward Overfitting and the Low-rank Dynamics in RLVR
par: Ye, Hao, et autres
Publié: (2026)

Detecting Generative Parroting through Overfitting Masked Autoencoders
par: Taghanaki, Saeid Asgari, et autres
Publié: (2024)

Test-time Adaptation for Regression by Subspace Alignment
par: Adachi, Kazuki, et autres
Publié: (2024)

LoRA Dropout as a Sparsity Regularizer for Overfitting Control
par: Lin, Yang, et autres
Publié: (2024)

Catastrophic Forgetting in Kolmogorov-Arnold Networks
par: Rahman, Mohammad Marufur, et autres
Publié: (2025)

Cross-Session Decoding of Neural Spiking Data via Task-Conditioned Latent Alignment
par: Zhao, Canyang, et autres
Publié: (2026)

Not All Timesteps Matter Equally: Selective Alignment Knowledge Distillation for Spiking Neural Networks
par: Sun, Kai, et autres
Publié: (2026)

Lying Is Just a Phase: The Hidden Alignment Transition in Language Model Scaling
par: Amin, Adil
Publié: (2026)

On the Privacy Risks of Spiking Neural Networks: A Membership Inference Analysis
par: Guan, Junyi, et autres
Publié: (2025)

Overcoming Overfitting in Reinforcement Learning via Gaussian Process Diffusion Policy
par: Horprasert, Amornyos, et autres
Publié: (2025)

From Overfitting to Reliability: Introducing the Hierarchical Approximate Bayesian Neural Network
par: Amirkhanian, Hayk, et autres
Publié: (2025)

Mitigating Overfitting in Graph Neural Networks via Feature and Hyperplane Perturbation
par: Choi, Yoonhyuk, et autres
Publié: (2022)

Transfer Learning for Benign Overfitting in High-Dimensional Linear Regression
par: Kim, Yeichan, et autres
Publié: (2025)