Hyperdimensional computing for structured querying on tabular data embeddings

2026-06-11

AI Summary

This work addresses the lack of interpretable similarity metrics in existing tabular embedding methods, which hinders reliable identification of queries with no matches. It introduces, for the first time, Holographic Reduced Representations (HRR) from Hyperdimensional Computing (HDC) into table row embeddings to support structured queries involving selection and projection. By leveraging the algebraic properties of HRR, the authors derive closed-form expressions for expected similarity under both equality and inequality predicates, enabling the establishment of interpretable and reliable retrieval thresholds. Experiments on two real-world datasets demonstrate that the proposed method outperforms the graph-based embedding baseline EmbDI, achieves perfect attribute projection accuracy in high-dimensional settings, and robustly handles inequality predicates while effectively detecting zero-match queries.

Abstract

Tabular data embeddings have become a cornerstone of data profiling and data integration pipelines, enabling tasks such as entity annotation and resolution; schema matching; column type detection; and table search, among others. Existing approaches embed rows, columns, or entire tables into a vector space and rely on nearest-neighbor search to retrieve candidate matches. A fundamental limitation of current embedding methods is the lack of interpretable similarity scores: the concrete similarity value between a query and its nearest neighbour carries no intrinsic meaning, making it impossible to determine whether that neighbour is a true match or simply the least-dissimilar item in a corpus that contains no valid answer. This inability to set principled thresholds for retrieval undermines practical deployment, particularly for zero-match detection. We investigate the use of HyperDimensional Computing (HDC), specifically the Holographic Reduced Representations (HRR) model, as a framework for tabular row embeddings when the retrieval task corresponds to answering structured select-project queries in vector space. Exploiting the algebraic properties of HDC operations, we derive closed-form expected similarity values for both equality and non-equality retrieval predicates, which converge to interpretable values as dimensionality increases, and use these to identify suitable retrieval thresholds. We evaluate HDC against EmbDI, a graph-based baseline, on two real-world datasets across varying table sizes and predicate lengths. Our results show that HDC matches or outperforms EmbDI for row retrieval across all configurations, handles non-equality predicates more robustly, and achieves perfect attribute projection accuracy at sufficient dimensionality -- while uniquely enabling reliable identification of zero-match predicates through its principled thresholds.

Problem

• tabular data embeddings
• interpretable similarity
• zero-match detection
• structured querying
• nearest-neighbor search

Innovation

• Hyperdimensional Computing
• Holographic Reduced Representations
• Tabular Data Embeddings
• Structured Querying
• Zero-Match Detection