Imagen de Portada

Grokking as a Variance-Limited Phase Transition: Spectral Gating and the Epsilon-Stability Threshold

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Acharya, Pratyush, Dhakal, Habish
Formato:	Preprint
Publicado:	2026
Materias:	Machine Learning Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Concepts Whisper While Syntax Shouts: Spectral Anti-Concentration and the Dual Geometry of Transformer Representations
por: Acharya, Pratyush, et al.
Publicado: (2026)

Grokking as a Falsifiable Finite-Size Transition
por: Bi, Yuda, et al.
Publicado: (2026)

Grokking as Dimensional Phase Transition in Neural Networks
por: Wang, Ping
Publicado: (2026)

Feature Repulsion and Spectral Lock-in: An Empirical Study of Two-Layer Network Grokking
por: Xu, Yongzhong
Publicado: (2026)

Grokking at the Edge of Numerical Stability
por: Prieto, Lucas, et al.
Publicado: (2025)

Topological Signatures of Grokking
por: Tang, Yifan, et al.
Publicado: (2026)

Dichotomy of Early and Late Phase Implicit Biases Can Provably Induce Grokking
por: Lyu, Kaifeng, et al.
Publicado: (2023)

Uncertainty Estimation using Variance-Gated Distributions
por: Gillis, H. Martin, et al.
Publicado: (2025)

The Geometric Inductive Bias of Grokking: Bypassing Phase Transitions via Architectural Topology
por: Yıldırım, Alper
Publicado: (2026)

The Geometry of Multi-Task Grokking: Transverse Instability, Superposition, and Weight Decay Phase Structure
por: Xu, Yongzhong
Publicado: (2026)

Spectral Gating Networks
por: Zhang, Jusheng, et al.
Publicado: (2026)

Stochastic Variance-Reduced Iterative Hard Thresholding in Graph Sparsity Optimization
por: Fox, Derek, et al.
Publicado: (2024)

Grokking in Linear Models for Logistic Regression
por: Das, Nataraj, et al.
Publicado: (2026)

Grokking Explained: A Statistical Phenomenon
por: Carvalho, Breno W., et al.
Publicado: (2025)

Controlling Grokking with Nonlinearity and Data Symmetry
por: Salah, Ahmed, et al.
Publicado: (2024)

Grokfast: Accelerated Grokking by Amplifying Slow Gradients
por: Lee, Jaerin, et al.
Publicado: (2024)

Understanding Grokking Through A Robustness Viewpoint
por: Tan, Zhiquan, et al.
Publicado: (2023)

Progress Measures for Grokking on Real-world Tasks
por: Golechha, Satvik
Publicado: (2024)

Low-Dimensional and Transversely Curved Optimization Dynamics in Grokking
por: Xu, Yongzhong
Publicado: (2026)

Tracing the Path to Grokking: Embeddings, Dropout, and Network Activation
por: Salah, Ahmed, et al.
Publicado: (2025)

The Geometry of Grokking: Norm Minimization on the Zero-Loss Manifold
por: Musat, Tiberiu
Publicado: (2025)

NeuralGrok: Accelerate Grokking by Neural Gradient Transformation
por: Zhou, Xinyu, et al.
Publicado: (2025)

Grokking Finite-Dimensional Algebra
por: Notsawo, Pascal Jr Tikeng, et al.
Publicado: (2026)

Grokking Group Multiplication with Cosets
por: Stander, Dashiell, et al.
Publicado: (2023)

Muon Optimizer Accelerates Grokking
por: Tveit, Amund, et al.
Publicado: (2025)

Spectral Entropy Collapse as a Phase Transition in Delayed Generalisation: An Interventional and Predictive Framework for Grokkin
por: Khanh, Truong Xuan, et al.
Publicado: (2026)

Early-Warning Signals of Grokking via Loss-Landscape Geometry
por: Xu, Yongzhong
Publicado: (2026)

Grokking as Structural Inference: Transformers Need Bayesian Lottery Tickets
por: Hidajat, Kai, et al.
Publicado: (2026)

Grokking and Generalization Collapse: Insights from \texttt{HTSR} theory
por: Prakash, Hari K., et al.
Publicado: (2025)

Spectral Edge Dynamics: An Analytical-Empirical Study of Phase Transitions in Neural Network Training
por: Xu, Yongzhong
Publicado: (2026)

From Density Matrices to Phase Transitions in Deep Learning: Spectral Early Warnings and Interpretability
por: Hennick, Max, et al.
Publicado: (2026)

Acceleration of Grokking in Learning Arithmetic Operations via Kolmogorov-Arnold Representation
por: Park, Yeachan, et al.
Publicado: (2024)

Provable Scaling Laws of Feature Emergence from Learning Dynamics of Grokking
por: Tian, Yuandong
Publicado: (2025)

Two Speeds of Learning: A Representation-Readout Decomposition of Grokking and Double Descent
por: Chou, Chi-Ning, et al.
Publicado: (2026)

Towards Empirical Interpretation of Internal Circuits and Properties in Grokked Transformers on Modular Polynomials
por: Furuta, Hiroki, et al.
Publicado: (2024)

The Norm-Separation Delay Law of Grokking: A First-Principles Theory of Delayed Generalization
por: Khanh, Truong Xuan, et al.
Publicado: (2026)

Anomaly Detection with Variance Stabilized Density Estimation
por: Rozner, Amit, et al.
Publicado: (2023)

ASGMamba: Adaptive Spectral Gating Mamba for Multivariate Time Series Forecasting
por: Li, Qianyang, et al.
Publicado: (2026)

Operationalizing Fairness: Post-Hoc Threshold Optimization Under Hard Resource Limits
por: Singh, Moirangthem Tiken, et al.
Publicado: (2026)

First-Passage Prediction of Grokking Delay: ACalibrated Law under AdamW with Causal Validation
por: Khanh, Truong Xuan, et al.
Publicado: (2026)