Yip, KH;
Changeat, Q;
Nikolaou, N;
Morvan, M;
Edwards, B;
Waldmann, IP;
Tinetti, G;
(2021)
Peeking inside the Black Box: Interpreting Deep-learning Models for Exoplanet Atmospheric Retrievals.
The Astronomical Journal
, 162
(5)
, Article 195. 10.3847/1538-3881/ac1744.
Preview |
Text
Yip_2021_AJ_162_195.pdf - Published Version Download (4MB) | Preview |
Abstract
Deep-learning algorithms are growing in popularity in the field of exoplanetary science due to their ability to model highly nonlinear relations and solve interesting problems in a data-driven manner. Several works have attempted to perform fast retrievals of atmospheric parameters with the use of machine-learning algorithms like deep neural networks (DNNs). Yet, despite their high predictive power, DNNs are also infamous for being "black boxes." It is their apparent lack of explainability that makes the astrophysics community reluctant to adopt them. What are their predictions based on? How confident should we be in them? When are they wrong, and how wrong can they be? In this work, we present a number of general evaluation methodologies that can be applied to any trained model and answer questions like these. In particular, we train three different popular DNN architectures to retrieve atmospheric parameters from exoplanet spectra and show that all three achieve good predictive performance. We then present an extensive analysis of the predictions of DNNs, which can inform us–among other things–of the credibility limits for atmospheric parameters for a given instrument and model. Finally, we perform a perturbation-based sensitivity analysis to identify to which features of the spectrum the outcome of the retrieval is most sensitive. We conclude that, for different molecules, the wavelength ranges to which the DNNs predictions are most sensitive do indeed coincide with their characteristic absorption regions. The methodologies presented in this work help to improve the evaluation of DNNs and to grant interpretability to their predictions.
| Type: | Article |
|---|---|
| Title: | Peeking inside the Black Box: Interpreting Deep-learning Models for Exoplanet Atmospheric Retrievals |
| Open access status: | An open access version is available from UCL Discovery |
| DOI: | 10.3847/1538-3881/ac1744 |
| Publisher version: | https://doi.org/10.3847/1538-3881/ac1744 |
| Language: | English |
| Additional information: | This version is the version of record. For information on re-use, please refer to the publisher's terms and conditions. |
| Keywords: | Convolutional neural networks (1938); Neural networks (1933); Exoplanet atmospheric composition (2021); Transit instruments (1708); Astronomical instrumentation (799) |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences > Dept of Physics and Astronomy |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10136651 |
Archive Staff Only
 |
View Item |
