Human Cognition-Inspired Hierarchical Fuzzy Learning Machine

Classification is a cornerstone of machine learning research. Most of the existing classifiers assume that the concepts corresponding to classes can be precisely defined. This notion diverges from the widely accepted understanding in cognitive science, which posits that real-world concepts are often inherently ambiguous. To bridge this big gap, we propose a Human Cognition-Inspired Hierarchical Fuzzy Learning Machine (HC-HFLM), which leverages a novel hierarchical alignment loss to integrate rich class knowledge from human knowledge system into learning process. We further theoretically prove that minimizing this loss can align the hierarchical structure derived from data with those contained in class knowledge, resulting in clear semantics and high interpretability. Systematic experiments verify that the proposed method can achieve significant gains in interpretability and generalization performance. Recently, we proposed a fuzzy learning machine (FLM) inspired by concept cognition (Cui & Liang, 2022) . FLM leverages fuzzy set theory (Zadeh, 1965) to capture the inherent fuzziness of concepts (McCloskey & Glucksberg, 1978; Marti et al., 2023) and employs exemplar theory to capture the typicality effects (Smith et al., 1974) , achieving good interpretability, robustness, and generalization. However, FLM still faces limitations in understanding con-