Physical Rule-Guided Convolutional Neural Network (2409.02081v1)
Abstract: 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.
- High-risk factor prediction in lung cancer using thin ct scans: An attention-enhanced graph convolutional network approach. In 2023 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), pages 1905–1910. IEEE, 2023.
- Generalizing graph convolutional neural networks with edge-variant recursions on graphs. In 2019 27th European Signal Processing Conference (EUSIPCO), pages 1–5. IEEE, 2019.
- Self-supervised learning of physics-guided reconstruction neural networks without fully sampled reference data. Magnetic resonance in medicine, 84(6):3172–3191, 2020.
- Ground-truth free multi-mask self-supervised physics-guided deep learning in highly accelerated mri. In 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI), pages 1850–1854. IEEE, 2021.
- Burhaneddin Yaman. Self-Supervised Physics-Guided Deep Learning for Solving Inverse Problems in Imaging. PhD thesis, University of Minnesota, 2022.
- Physics-guided deep neural network for structural damage identification. Ocean Engineering, 260:112073, 2022.
- Sentence and text comprehension: Roles of linguistic structure. Annual Review of Psychology, 52(1):167–196, 2001.
- Douglas Biber. Text-linguistic approaches to register variation. Register studies, 1(1):42–75, 2019.
- Effects of pitch and speech rate on personal attributions. Journal of personality and social psychology, 37(5):715, 1979.
- Pgnn: Physics-guided neural network for fourier ptychographic microscopy. arXiv preprint arXiv:1909.08869, 2019.
- Physics-informed computer vision: A review and perspectives. arXiv preprint arXiv:2305.18035, 2023.
- Physics-guided neural networks for intraventricular vector flow mapping. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, 2024.
- Physics-guided neural network model for aeroengine control system sensor fault diagnosis under dynamic conditions. Aerospace, 10(7):644, 2023.
- Physics-guided bayesian neural networks by abc-ss: Application to reinforced concrete columns. Engineering Applications of Artificial Intelligence, 119:105790, 2023.
- Physics guided remote sensing image synthesis network for ship detection. IEEE Transactions on Geoscience and Remote Sensing, 61:1–14, 2023.
- Dock: Detecting objects by transferring common-sense knowledge. In Proceedings of the European Conference on Computer Vision (ECCV), pages 492–508, 2018.
- VilniusTech University. Cars from drone dataset. https://universe.roboflow.com/vilniustech-university/cars-from-drone-f5kny, nov 2023. visited on 2024-08-02.
- Mranmay Shetty. Dronevehicle dataset. https://universe.roboflow.com/mranmay-shetty/dronevehicle, may 2023. visited on 2024-08-13.
- Dataset Ninja. Visualization tools for ship detection from aerial images dataset. https://datasetninja.com/aerial-ship-detection, aug 2024. visited on 2024-08-02.
- U-net: Convolutional networks for biomedical image segmentation. In Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part III 18, pages 234–241. Springer, 2015.
- U2-net: Going deeper with nested u-structure for salient object detection. Pattern recognition, 106:107404, 2020.
- Segment anything. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 4015–4026, 2023.
Collections
Sign up for free to add this paper to one or more collections.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.