Saved in:
Bibliographic Details
Main Authors: Wu, Dennis, Hu, Jerry Yao-Chieh, Hsiao, Teng-Yun, Liu, Han
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2404.03827
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912112524132352
author Wu, Dennis
Hu, Jerry Yao-Chieh
Hsiao, Teng-Yun
Liu, Han
author_facet Wu, Dennis
Hu, Jerry Yao-Chieh
Hsiao, Teng-Yun
Liu, Han
contents We propose a two-stage memory retrieval dynamics for modern Hopfield models, termed $\mathtt{U\text{-}Hop}$, with enhanced memory capacity. Our key contribution is a learnable feature map $Φ$ which transforms the Hopfield energy function into kernel space. This transformation ensures convergence between the local minima of energy and the fixed points of retrieval dynamics within the kernel space. Consequently, the kernel norm induced by $Φ$ serves as a novel similarity measure. It utilizes the stored memory patterns as learning data to enhance memory capacity across all modern Hopfield models. Specifically, we accomplish this by constructing a separation loss $\mathcal{L}_Φ$ that separates the local minima of kernelized energy by separating stored memory patterns in kernel space. Methodologically, $\mathtt{U\text{-}Hop}$ memory retrieval process consists of: (Stage I) minimizing separation loss for a more uniform memory (local minimum) distribution, followed by (Stage II) standard Hopfield energy minimization for memory retrieval. This results in a significant reduction of possible metastable states in the Hopfield energy function, thus enhancing memory capacity by preventing memory confusion. Empirically, with real-world datasets, we demonstrate that $\mathtt{U\text{-}Hop}$ outperforms all existing modern Hopfield models and state-of-the-art similarity measures, achieving substantial improvements in both associative memory retrieval and deep learning tasks. Code is available at https://github.com/MAGICS-LAB/UHop ; future updates are on arXiv:2404.03827
format Preprint
id arxiv_https___arxiv_org_abs_2404_03827
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Uniform Memory Retrieval with Larger Capacity for Modern Hopfield Models
Wu, Dennis
Hu, Jerry Yao-Chieh
Hsiao, Teng-Yun
Liu, Han
Machine Learning
Artificial Intelligence
We propose a two-stage memory retrieval dynamics for modern Hopfield models, termed $\mathtt{U\text{-}Hop}$, with enhanced memory capacity. Our key contribution is a learnable feature map $Φ$ which transforms the Hopfield energy function into kernel space. This transformation ensures convergence between the local minima of energy and the fixed points of retrieval dynamics within the kernel space. Consequently, the kernel norm induced by $Φ$ serves as a novel similarity measure. It utilizes the stored memory patterns as learning data to enhance memory capacity across all modern Hopfield models. Specifically, we accomplish this by constructing a separation loss $\mathcal{L}_Φ$ that separates the local minima of kernelized energy by separating stored memory patterns in kernel space. Methodologically, $\mathtt{U\text{-}Hop}$ memory retrieval process consists of: (Stage I) minimizing separation loss for a more uniform memory (local minimum) distribution, followed by (Stage II) standard Hopfield energy minimization for memory retrieval. This results in a significant reduction of possible metastable states in the Hopfield energy function, thus enhancing memory capacity by preventing memory confusion. Empirically, with real-world datasets, we demonstrate that $\mathtt{U\text{-}Hop}$ outperforms all existing modern Hopfield models and state-of-the-art similarity measures, achieving substantial improvements in both associative memory retrieval and deep learning tasks. Code is available at https://github.com/MAGICS-LAB/UHop ; future updates are on arXiv:2404.03827
title Uniform Memory Retrieval with Larger Capacity for Modern Hopfield Models
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2404.03827