EXO-INSIGHT: An Explainable Multi-Modal Deep Learning Framework for Enhanced Exoplanet Detection

References

[1] W. J. Borucki et al., “Kepler planet-detection mission: Introduction and first results,” Science, vol. 327, no. 5968, pp. 977–980, 2010.

[2] G. R. Ricker et al., “Transiting Exoplanet Survey Satellite (TESS),” Journal of Astronomical Telescopes, Instruments, and Systems, vol. 1, no. 1, pp. 014003, 2015.

[3] J. N. Winn and D. C. Fabrycky, “The occurrence and architecture of exoplanetary systems,” Annual Review of Astronomy and Astrophysics, vol. 53, pp. 409–447, 2015.

[4] D. Shallue and A. Vanderburg, “Identifying exoplanets with deep learn-ing: A five-planet resonant chain around Kepler-80 and an eighth planet around Kepler-90,” The Astronomical Journal, vol. 155, no. 2, 2018.

[5] T. Barclay, J. Pepper, and S. Quintana, “A revised exoplanet yield from the TESS primary mission,” The Astrophysical Journal Supplement Series, vol. 239, no. 1, 2018.

[6] A. Pearson, S. Palafox, and L. Griffith, “Searching for exoplanets using artificial intelligence,” Monthly Notices of the Royal Astronomical Society, vol. 474, no. 1, pp. 478–491, 2018.

[7] Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521,pp. 436–444, 2015.

[8] A. Krizhevsky, I. Sutskever, and G. Hinton, “ImageNet classification with deep convolutional neural networks,” in Advances in Neural Infor-mation Processing Systems, pp. 1097–1105, 2012.

[9] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition, pp. 770–778, 2016.

[10] J. Schmidhuber, “Deep learning in neural networks: An overview,” Neural Networks, vol. 61, pp. 85–117, 2015.

[11] S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997.

[12] F. Chollet, Deep Learning with Python. Manning Publications, 2017.

[13] R. R. Selvaraju et al., “Grad-CAM: Visual explanations from deep networks via gradient- based localization,” in Proc. IEEE International Conference on Computer Vision, pp. 618–626, 2017.

[14] Z. C. Lipton, “The mythos of model interpretability,” Communications of the ACM, vol. 61, no. 10, pp. 36–43, 2018.

[15] N. M. Batalha et al., “Planetary candidates observed by Kepler,” The Astrophysical Journal Supplement Series, vol. 204, no. 2, 2013.

[16] T. D. Morton et al., “False positive probabilities for all Kepler objects of interest,” The Astrophysical Journal, vol. 822, no. 2, 2016.

[17] J. K. Becker et al., “Machine learning for exoplanet transit detection in space-based surveys,” Astronomy and Computing, vol. 25, pp. 1–9, 2018.

[18] E. J. Armstrong et al., “Transit shapes and self-organizing maps as a tool for ranking planetary candidates,” Monthly Notices of the Royal Astronomical Society, vol. 465, no. 3, pp. 2634–2642, 2017.

[19] A. Esteva et al., “A guide to deep learning in healthcare,” Nature Medicine, vol. 25, no. 1, pp. 24–29, 2019.

[20] K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” in Proc. International Conference on Learning Representations (ICLR), 2015.

[21] T. N. Kipf and M. Welling, “Semi-supervised classification with graph convolutional networks,” in Proc. International Conference on Learning Representations (ICLR), 2017.

@article{1715145,
      author = {Issac Basil Eldho, Joel Joji, KJ Allwyn, Hiran Jebi},
      title = {EXO-INSIGHT: An Explainable Multi-Modal Deep Learning Framework for Enhanced Exoplanet Detection},
      journal = {Iconic Research And Engineering Journals},
      year = {2026},
      volume = {9},
      number = {9},
      pages = {1816-1824},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1715145.pdf},
      abstract = {Because of the rapid increase in astronomical ob-servation missions, a huge amount of data is generated, which should be analyzed to detect exoplanets that may be present in the data. Space-based telescopes such as the Kepler mission and the Transiting Exoplanet Survey Satellite (TESS) have generated large amounts of data in the form of photometric light curves and astronomical data [1], [2]. The current study introduces an explainable multi-modal deep learning framework named EXO-INSIGHT to improve exoplanet detection by analyzing temporal and spatial astronomical data. The framework analyzes the changes in the brightness of stars from light curve sequences using temporal learning models, whereas it also analyzes astro-nomical images to extract features using CNNs. The features from these two modalities are combined using a feature fusion mechanism to improve the reliability of the exoplanet detection process. Moreover, the framework also includes explainability techniques to highlight the relevant region of images and relevant segments of light curves to provide insights into the detection process [13]. The framework is implemented in the form of a web-based platform to provide researchers with a facility to upload their data and obtain results from the framework with explanations for the prediction results provided by the framework. The framework is expected to provide reliable results for analyzing large amounts of astronomical data for exoplanet detection.},
      keywords = {Exoplanet Detection, Multi-Modal Deep Learning, Explainable Artificial Intelligence, Astronomical Data Analysis, Light Curve Analysis, Convolutional Neural Networks, Feature Fusion, Space Telescope Data.},
      month = {March},
      doi = {https://doi.org/10.64388/IREV9I9-1715145}
  }