arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~

学习克利福德变分自编码器的全息约简表示

Learning Holographic Reduced Representations with Clifford Variational Autoencoders

Mohamed Malek Abid, P. Michael Furlong

arXiv 2609.28409首次发表:更新:

发表机构

Institute of Neuroinformatics; University of Zurich & ETH Zurich; National Research Council Canada; University of Waterloo(神经信息学研究所; 苏黎世大学与苏黎世联邦理工学院; 加拿大国家研究委员会; 滑铁卢大学)

机构由 AI 辅助整理,请以论文原文为准。

AI 中文总结

本文提出Clifford-VAE,一种将数据投影到任意维度克利福德环面的变分自编码器,用于学习全息约简表示,在半监督分类中与高斯和超球面VAE竞争,并在VSA基准测试中表现更优,为感知数据接地到符号推理提供了新方法。

AI 中文摘要

向量符号代数通过将其向量代数应用于随机生成的原子向量符号和实值数据的分数幂编码,将数据结构投影到超维向量空间中。嵌入非结构化数据仍然是一个悬而未决的问题。我们提出了\textit{Clifford-VAE},一种变分自编码器,它学习将数据投影到任意维度的克利福德环面上。使用MNIST、FashionMNIST和CIFAR-10数据集的实验表明,在半监督分类任务中,Clifford-VAE产生的表示与高斯和超球面VAE产生的表示具有竞争力,同时在VSA基准测试中,在自绑定和解绑定、角色-填充恢复以及捆绑容量方面优于高斯和超球面VAE。Clifford-VAE提供了一种将感知数据接地到符号推理框架中的原则性技术,为VSA文献中一个长期存在的问题提供了一种新方法。

英文摘要

Vector Symbolic Algebras project data structures into a hyperdimensional vector space through the application of their vector algebras to randomly generated atomic vector symbols and fractional power encodings of real-valued data. Embedding unstructured data remains an open question. We present \textit{Clifford-VAE}, a variational autoencoder that learns to project data onto a Clifford torus in arbitrary dimensions. Experiments using the MNIST, FashionMNIST, and CIFAR-10 datasets demonstrate that Clifford-VAE produces representations that are competitive with those produced by Gaussian and Hyperspherical VAEs for semi-supervised classification tasks while outperforming Gaussian and Hyperspherical counterparts in the VSA benchmark tests of self-binding and unbinding, role-filler recovery, and bundle capacity. Clifford-VAE provides a principled technique for grounding perceptual data into a symbolic reasoning framework, providing a new approach to a long-standing problem in the VSA literature.

CommentsPreprint. 24 pages, 20 figures

论文原文

arXiv 摘要页 · PDF 原文 · HTML 原文

↑