Physics-Informed Spectral Modeling for Hyperspectral Imaging
Not on the map. Why?
Abstract (arXiv)
We present PhISM, a physics-informed deep learning architecture that learns without supervision to explicitly disentangle hyperspectral observations and model them with continuous basis functions. PhISM outperforms prior methods on several classification and regression benchmarks, requires limited labeled data, and provides additional insights thanks to interpretable latent representation.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 11 of its 40 references that are in the map. First posted on arXiv 29 Aug 2025.
Builds on
11 of 40 references.
References · 11
- 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,166
- 2019Optuna: A Next-generation Hyperparameter Optimization Frameworkcited by 2,501
- 2021SpectralFormer: Rethinking Hyperspectral Image Classification with Transformerscited by 169
- 2019Neural-Symbolic Computing: An Effective Methodology for Principled Integration of Machine Learning and Reasoningcited by 151
- 2021Incorporating Domain Knowledge into Deep Neural Networkscited by 56
- 2024shapiq: Shapley Interactions for Machine Learningcited by 43
- 2018Validating Hyperspectral Image Segmentationcited by 18
- 2021Model Inspired Autoencoder for Unsupervised Hyperspectral Image Super-Resolutioncited by 10
- 2016BASS Net: Band-Adaptive Spectral-Spatial Feature Learning Neural Network for Hyperspectral Image Classificationcited by 5
- 2024Physics-informed Machine Learning for Battery Pack Thermal Managementcited by 2
- 2025Tobler's First Law in GeoAI: A Spatially Explicit Deep Learning Model for Terrain Feature Detection Under Weak Supervisioncited by 1
Built on by
Citing papers per year: 2026: 2.
2 papers.