Hello! 
I’d like to discuss an idea of uncertain scope and feasibility.
So, I’m interested in working on a deep learning project, and I have this idea about training a Spiking Neural Network (SNN) to analyze subtle spectral data from exoplanets’ biosignatures.
However, I’m unsure about the effectiveness of SNNs compared to CNNs (or even RNNs or a hybrid of both) for this specific task, especially given the limited studies and experiments.
I’ve already asked some professionals who have worked on similar projects, and they too are unsure about the whole thing.
Has anyone here explored the application of SNNs in analyzing spectral data or have insights on how to approach this? I’d appreciate any guidance or commentary.
Take care, everyone. 
Peppermintbird,
I have not worked on SNN’s. I did find this website: Exoplanet Hunter: Finding Planets Using Light Intensity — snntorch 0.7.0 documentation. It might be interesting to read if you have not already run across it. However, I could not find anything with the biosignatures angle. I wish you luck!
Katherine Moss
Hi Katherine!
Thank you very much for the message. I study snnTorch (a Python package using SNNs) and try to collaborate with its open source. Coincidence? Still, thank you for looking into it. 
The thing is, in the project you sent, snnTorch didn’t use the temporal axis of the SNNs, which is one of the reasons I’m interested in this neural network for analyzing exoplanetary data. Sadly, there aren’t many studies into it.
1 Like