Deep Tree Tensor Networks for Image Recognition

• DOI: 10.48550/arxiv.2502.09928
• Publication Date: 2025-02-14
• Authors: Chang Nie; Junfang Chen; Yajie Chen
• Keywords: Tree (set theory)

Originating in quantum physics, tensor networks (TNs) have been widely adopted as exponential machines and parameter decomposers for recognition tasks. Typical TN models, such Matrix Product States (MPS), not yet achieved successful application natural image processing. When employed, they primarily serve to compress parameters within off-the-shelf networks, thus losing their distinctive capability enhance exponential-order feature interactions. This paper introduces a novel architecture named \textit{\textbf{D}eep \textbf{T}ree \textbf{T}ensor \textbf{N}etwork} (DTTN), which captures $2^L$-order multiplicative interactions across features through multilinear operations, while essentially unfolding into \emph{tree}-like topology with the parameter-sharing property. DTTN is stacked multiple antisymmetric interacting modules (AIMs), this design facilitates efficient implementation. Moreover, we theoretically reveal equivalency among quantum-inspired models polynomial under certain conditions, believe that can inspire more interpretable studies field. We evaluate proposed model against series of benchmarks achieve excellent performance compared its peers cutting-edge architectures. Our code will soon be publicly available.