Knowledge-Inclusive Adaptive Physics-Informed Neural Network for Microbial Interaction Modelling

We propose a knowledge-inclusive adaptive physics-informed neural network (PINN) for modelling microbial interactions. By embedding domain knowledge and governing dynamics into the learning process, the approach improves the modelling of complex microbial community interactions beyond purely data-driven methods, while adapting to the characteristics of the underlying system.

Download paper here

Recommended citation:

R. Rupasinghe, R. Vidanaarachchi, A. Hevapathige, S. Seneviratne, S.-L. Tang and S. Halgamuge, “Knowledge-Inclusive Adaptive Physics-Informed Neural Network for Microbial Interaction Modelling,” arXiv preprint arXiv:2606.07686, 2026.

BibTeX

@article{rupasinghe2026knowledge,
  title={Knowledge-Inclusive Adaptive Physics-Informed Neural Network for Microbial Interaction Modelling},
  author={Rupasinghe, Ravisha and Vidanaarachchi, Rajith and Hevapathige, Asela and Seneviratne, Sachith and Tang, Sen-Lin and Halgamuge, Saman},
  journal={arXiv preprint arXiv:2606.07686},
  year={2026}
}