paper-with-me

Papers

The Complexity Dynamics of Grokking

2024-12-13 · Branton DeMoss, Silvia Sapora, Jakob Foerster, Nick Hawes, Ingmar Posner

We investigate the phenomenon of generalization through the lens of compression. In particular, we study the complexity dynamics of neural networks to explain grokking, where networks suddenly transition from memorizing to generalizing solutions long after over-fitting the training data. To this end we introduce a new measure of intrinsic complexity for neural networks based on the theory of Kolmogorov complexity. Tracking this metric throughout network training, we find a consistent pattern in training dynamics, consisting of a rise and fall in complexity. We demonstrate that this corresponds to memorization followed by generalization. Based on insights from rate--distortion theory and the minimum description length principle, we lay out a principled approach to lossy compression of neural networks, and connect our complexity measure to explicit generalization bounds. Based on a careful analysis of information capacity in neural networks, we propose a new regularization method which encourages networks towards low-rank representations by penalizing their spectral entropy, and find that our regularizer outperforms baselines in total compression of the dataset.

📄 PDF Abstract BibTeX arXiv:2412.09810

Code (1)

brantondemoss/GrokkingComplexity 공식 구현 jax

Tasks

Generalization BoundsMemorization

Similar Papers 제목 키워드 기반

Progress Measures for Grokking on Real-world Tasks

2024-05-21 · Satvik Golechha

Grokking, a phenomenon where machine learning models generalize long after overfitting, has been primarily observed and studied in algorithmic tasks. This paper explores grokking in real-world datasets using deep neural …

How to explain grokking

2024-12-17 · S. V. Kozyrev

Explanation of grokking (delayed generalization) in learning is given by modeling grokking by the stochastic gradient Langevin dynamics (Brownian motion) and applying the ideas of thermodynamics.

Where to find Grokking in LLM Pretraining? Monitor Memorization-to-Generalization without Test

2025-06-26 · Ziyue Li, Chenrui Fan, Tianyi Zhou

Grokking, i.e., test performance keeps improving long after training loss converged, has been recently witnessed in neural network training, making the mechanism of generalization and other emerging capabilities such as …

Code GenerationLarge Language ModelMathMemorization

Algorithmic Information Dynamics of Learning: A Certified, Differentiable Complexity Controller for Grokking

2026-08-14 · Luan Ozelim, Abicumaran Uthamacumaran, Hector Zenil arxiv

Algorithmic Information Dynamics (AID) studies systems by perturbing them and measuring changes in algorithmic complexity, but its usual estimator, the Block Decomposition Method, is piecewise constant, restricting the c…

Grokking Beyond Neural Networks: An Empirical Exploration with Model Complexity

2023-10-26 · Jack Miller, Charles O'Neill, Thang Bui

In some settings neural networks exhibit a phenomenon known as \textit{grokking}, where they achieve perfect or near-perfect accuracy on the validation set long after the same performance has been achieved on the trainin…

regression