HQFNN: A Compact Quantum-Fuzzy Neural Network for Accurate Image Classification
This work provides a compact, interpretable, and noise-tolerant alternative to conventional vision backbones, addressing issues in image classification for applications requiring robustness and transparency.
The paper tackled the problem of noisy inputs and lack of interpretability in deep learning vision systems by introducing a Highly Quantized Fuzzy Neural Network (HQFNN), which achieved higher accuracy than classical, fuzzy-enhanced, and quantum-only baselines on standard image benchmarks while using orders of magnitude fewer trainable weights and showing marginal accuracy degradation under simulated noise.
Deep learning vision systems excel at pattern recognition yet falter when inputs are noisy or the model must explain its own confidence. Fuzzy inference, with its graded memberships and rule transparency, offers a remedy, while parameterized quantum circuits can embed features in richly entangled Hilbert spaces with striking parameter efficiency. Bridging these ideas, this study introduces a innovative Highly Quantized Fuzzy Neural Network (HQFNN) that realises the entire fuzzy pipeline inside a shallow quantum circuit and couples the resulting quantum signal to a lightweight CNN feature extractor. Each image feature is first mapped to a single qubit membership state through repeated angle reuploading. Then a compact rule layer refines these amplitudes, and a clustered CNOT defuzzifier collapses them into one crisp value that is fused with classical features before classification. Evaluated on standard image benchmarks, HQFNN consistently surpasses classical, fuzzy enhanced and quantum only baselines while using several orders of magnitude fewer trainable weights, and its accuracy degrades only marginally under simulated depolarizing and amplitude damping noise, evidence of intrinsic robustness. Gate count analysis further shows that circuit depth grows sublinearly with input dimension, confirming the model's practicality for larger images. These results position the model as a compact, interpretable and noise tolerant alternative to conventional vision backbones and provide a template for future quantum native fuzzy learning frameworks.