Physical Rule-Guided Convolutional Neural Network
This addresses the problem of improving CNN interpretability and efficiency for complex domains with limited labeled data, representing an incremental advancement in physics-guided neural networks.
The paper tackles the limitations of CNNs in data-scarce domains by proposing a Physics-Guided CNN (PGCNN) that integrates LLM-generated rules as custom layers, achieving superior performance with reduced false positives and enhanced confidence scores compared to baseline CNNs.
The black-box nature of Convolutional Neural Networks (CNNs) and their reliance on large datasets limit their use in complex domains with limited labeled data. Physics-Guided Neural Networks (PGNNs) have emerged to address these limitations by integrating scientific principles and real-world knowledge, enhancing model interpretability and efficiency. This paper proposes a novel Physics-Guided CNN (PGCNN) architecture that incorporates dynamic, trainable, and automated LLM-generated, widely recognized rules integrated into the model as custom layers to address challenges like limited data and low confidence scores. The PGCNN is evaluated on multiple datasets, demonstrating superior performance compared to a baseline CNN model. Key improvements include a significant reduction in false positives and enhanced confidence scores for true detection. The results highlight the potential of PGCNNs to improve CNN performance for broader application areas.