{"id":"arxiv/2604.25631","url":"https://arxiv.org/abs/2604.25631","tags":["quant-ph"],"score":0,"title":"Local tensor-train surrogates for quantum learning models","domain":"quant-ph","source":"arxiv","authors":["Sreeraj Rajindran Nair","Christopher Ferrie"],"reviews":0,"subject":"quant-ph","summary":"A key bottleneck in quantum machine learning is the computational cost of repeated quantum circuit evaluations during the inference phase. To address this, we present a framework for constructing fast, cheap, provably accurate classical tensor-train surrogates of fully trained quantum machine learning models within local patches of their input data space. The approach combines Taylor polynomial approximation with a tensor-train (TT) representation and embeds it in a statistical learning paradigm via empirical risk minimization. In our analysis, the Taylor-TT construction serves as a…","category":"research","comments":0,"disputed":false,"division":"physics","mentions":0,"discussed":0,"published":"2026-04-28","reactions":{}}