The prediction of sensory attributes from ingredient-level formulations is an emerging challenge at the intersection of food science and artificial intelligence. We address the fundamental question of whether the taste of a food can be predicted from its ingredients by treating recipes as composite materials. We apply Hashin--Shtrikman (HS) and Reuss--Voigt (RV) bounds, techniques originally developed for elastic moduli, to predict five taste dimensions (sweetness, sourness, bitterness, umami, saltiness) on a curated dataset of 70 recipes decomposed into 209 ingredient-level taste references with trained-panel ground truth. The bounds provided an additive baseline but systematically under-predict perceived taste: 77\% of actual taste values exceeded the HS upper bound, with the exceedance rate ranging from 26\% (bitterness) to 97\% (saltiness). We traced this gap to specific processing chemistry (Maillard reactions, caramelization, evaporative concentration, protein hydrolysis, and nucleotide synergy) and introduced a hybrid model that augments the HS baseline with eight chemistry-proxy features encoding these mechanisms. Our results show that our interpretable hybrid model eliminates the systematic bias and reduces mean absolute error by 27--62\% for sweetness, sourness, umami, and saltiness while using only 10 interpretable features, achieving performance comparable to a black-box Lasso regression on 115 per-ingredient features. We further demonstrate constrained inverse design via Differential Evolution, recovering ingredient formulations that match target taste profiles subject to compositional bounds.

翻译：从配方成分级别预测感官属性是食品科学与人工智能交叉领域的新兴挑战。我们通过将食谱视为复合材料，探讨能否从成分预测食品口味这一基本问题。采用原本用于弹性模量预测的Hashin-Shtrikman（HS）和Reuss-Voigt（RV）界，在包含70个食谱（分解为209个成分级口味参考值）并配有训练专家小组真实数据的精选数据集上，预测五个口味维度（甜味、酸味、苦味、鲜味、咸味）。这些界提供了可加性基准，但系统性地低估了感知口味：77%的实际口味值超过HS上界，超出率从26%（苦味）到97%（咸味）不等。我们将此偏差归因于特定加工化学过程（美拉德反应、焦糖化、蒸发浓缩、蛋白质水解及核苷酸协同作用），并引入混合模型——通过八个编码这些机制的化学代理特征增强HS基准。结果表明，我们的可解释混合模型消除了系统性偏差，仅使用10个可解释特征就将甜味、酸味、鲜味和咸味的平均绝对误差降低27-62%，其性能可与使用115个成分特征的不可解释Lasso回归相媲美。我们进一步通过差分进化展示受约束的逆向设计，实现在成分界约束下恢复与目标口味特征匹配的配方组成。