BriefGPT.xyz
Oct, 2024
通过随机特征观察稠密关联记忆
Dense Associative Memory Through the Lens of Random Features
HTML
PDF
Benjamin Hoover, Duen Horng Chau, Hendrik Strobelt, Parikshit Ram, Dmitry Krotov
TL;DR
本研究解决了稠密关联记忆在引入新记忆模式时需要增加突触权重数量的问题。我们提出了一种使用随机特征的替代模型,保持网络参数数量不变,同时允许通过调整现有权重来增加新记忆。研究结果表明,该模型可以紧密近似传统稠密关联记忆的能量函数和动态特性,并共享其优良的计算特性。
Abstract
Dense Associative Memories are high storage capacity variants of the
Hopfield Networks
that are capable of storing a large number of
Memory Patterns
in the weights of the network of a given size. Their common for
→