XClose

The Centre for Planetary Sciences at UCL / Birkbeck

Home
Menu

Special seminar: Exoplanets exploration in the era of 'Big Data' and AI

29 March 2023, 4:00 pm–5:00 pm

Ariel Data Challenge logo

Professor Katia Matcheva, representing the winning team of the Ariel Data Challenge 2022, will give a special seminar on "Exoplanets Exploration in the Era of 'Big Data' and AI".

This event is free.

Event Information

Open to

All

Availability

Yes

Cost

Free

Organiser

Gordon Yip

On Wednesday 29th March, a special seminar will be given by Professor Katia Matcheva, representing the winning team of the Ariel Data Challenge 2022, on the topic of "Exoplanets Exploration in the Era of 'Big Data' and AI".
 
Prof Matcheva will give her talk via Zoom, but a room has also been reserved for people who would like to get together to watch it.
 
Location: UCL North-West Wing Lecture Theatre G22
Date: Wednesday 29 March 2023
Time: 4 – 5 PM

Please contact the organiser for Zoom joining details. 


Abstract

Transmission spectroscopy is a powerful tool to decode the chemical composition of the atmospheres of transiting extrasolar planets. Our ability to reliably and meaningfully extract information about their physical structure and chemical composition from the observed spectra relies on the complete understanding of the mathematics of the problem, knowing the uncertainties of the observations, as well as, on our ability to handle large volumes of data associated with high-resolution spectra in the era of large planetary surveys.  
 
In this talk I will present the winning machine learning solution to the ARIEL 2022 Big Data Challenge, which was submitted by the University of Florida team, “Gators”, and will discuss several lines of theoretical and machine learning inquiries, which our planetary group is currently working on. 
 
About the "Gators" team:
We are a research team from the University of Florida with diverse fields of expertise encompassing planetary astronomy, elementary particle theory, inferential statistics, Monte Carlo simulations, high-performance computing, and machine learning. Our favorite method is “intellectual arbitrage”, i.e., leveraging ideas from one scientific community to solve outstanding problems in another. 

Links