paper-with-me

Papers

Grokking Modular Polynomials

2024-06-05 · Darshil Doshi, Tianyu He, Aritra Das, Andrey Gromov

Neural networks readily learn a subset of the modular arithmetic tasks, while failing to generalize on the rest. This limitation remains unmoved by the choice of architecture and training strategies. On the other hand, an analytical solution for the weights of Multi-layer Perceptron (MLP) networks that generalize on the modular addition task is known in the literature. In this work, we (i) extend the class of analytical solutions to include modular multiplication as well as modular addition with many terms. Additionally, we show that real networks trained on these datasets learn similar solutions upon generalization (grokking). (ii) We combine these "expert" solutions to construct networks that generalize on arbitrary modular polynomials. (iii) We hypothesize a classification of modular polynomials into learnable and non-learnable via neural networks training; and provide experimental evidence supporting our claims.

📄 PDF Abstract BibTeX arXiv:2406.03495

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Towards Empirical Interpretation of Internal Circuits and Properties in Grokked Transformers on Modular Polynomials

2024-02-26 · Hiroki Furuta, Gouki Minegishi, Yusuke Iwasawa, Yutaka Matsuo

Grokking has been actively explored to reveal the mystery of delayed generalization and identifying interpretable representations and algorithms inside the grokked models is a suggestive hint to understanding its mechani…

Grokking modular arithmetic

2023-01-06 · Andrey Gromov

We present a simple neural network that can learn modular arithmetic tasks and exhibits a sudden jump in generalization known as ``grokking''. Concretely, we present (i) fully-connected two-layer networks that exhibit gr…

Model Capacity Determines Grokking through Competing Memorisation and Generalisation Speeds

2026-05-10 · Yiding Song, Hanming Ye arxiv

Existing accounts of grokking explain the phenomena in terms of mechanistic frameworks such as circuit efficiency or lazy-to-rich transitions. However, despite a known dependence between grokking and model size, how mode…

Circuit Synchronization Precedes Generalization: A Causal Precursor to Grokking

2026-06-11 · Achyuthan Sivasankar arxiv

Grokking is the delayed generalisation phenomenon where a transformer trained on modular arithmetic abruptly transitions from near-chance to near-perfect validation accuracy. It has been attributed to a Fourier-based alg…

Why Do You Grok? A Theoretical Analysis of Grokking Modular Addition

2024-07-17 · Mohamad Amin Mohamadi, Zhiyuan Li, Lei Wu, Danica J. Sutherland

We present a theoretical explanation of the ``grokking'' phenomenon, where a model generalizes long after overfitting,for the originally-studied problem of modular addition. First, we show that early in gradient descent,…