Artash Nath The Americas World Summit on Artificial Intelligence was held on 25-26 March 2020. Due to the ongoing COVID19 pandemic, the conference organizers took the correct step of shifting […]
The Americas World Summit on Artificial Intelligence was held on 25-26 March 2020. Due to the ongoing COVID19 pandemic, the conference organizers took the correct step of shifting the conference location from Palais des congrès de Montréal, Canada to a virtual event.
The summit was run over an on-line platform over the two days and all the content, including presentations, talks, and panel discussions were made available in a pre-recorded format through the World Summit on Artificial Intelligence (WSAI) TV. It was a wonderful opportunity to listen to other presenters from all over the world and be able to make connections with other participants.
Presenting at the Americas World Summit on Artificial Intelligence
As I had to deliver my talk from remote, I created a fresh presentation using slides, visuals, and props specifically for the audience of the World Summit on AI and recorded it on video. I realized that while many of the participants of the conference may be knowledgeable about Artificial Intelligence, they may not be familiar with the thematic application of my talk, namely the science of exoplanets discovery and how to predict the chemical composition of their atmospheres.
To excite the audience about my talk and how machine learning was applied to it, I had earlier given an interview to the organizers of the World Summit on Artificial Intelligence “Are We Alone in the Universe” which was published on their blog at https://blog.worldsummit.ai/are-we-alone-in-the-universe
Using Machine Learning on Data from the ARIEL Space Telescope
My talk focused on simulated data obtained from the Atmospheric Remote-sensing Infrared Exoplanet Large-survey (ARIEL) space telescope. The ARIEL Space Telescope is the fourth medium-class mission in the European Space Agency’s Cosmic Vision program that will study what exoplanets are made of, how they formed and how they evolve, by surveying a diverse sample of about 1000 extrasolar planets, simultaneously in visible and infrared wavelengths. It will be measuring the chemical composition and thermal structures of hundreds of transiting exoplanets providing new insights on planetary science beyond our solar system.
Before researchers can analyze the data it was important to remove noise from exoplanet observations caused by starspots and by instrumentation in the simulated data from the ARIEL telescope. And that is where Machine Learning played a very important role. Machine Learning is very useful in identifying patterns in large data sets and handle different data types.
In the case of the ARIEL Space Telescope, over 150,000 simulated observations were made available alongside transit light curves in 55 different wavelengths and 300 time-step data points. In addition, 6 stellar parameters and planet-star radius ratios were provided. It took a lot of time to pre-process the data, rearrange it, divide it into training and testing data before running different machine learning models to determine the most accurate one.
The model which I finally came up with was a hybrid machine learning model. The Model uses the Long Short Term Memory (LSTM) Model, a form of Recurrent Neural Network (RNN) to handle the time series (or sequential) data such as transit light curves. It uses the Feed-Forward Neural Network to handle the numerical data such as mass, radius, temperatures, period, and the magnitude of stars generated by ARIEL. I then applied the Concatenate Layer to merge the two machine learning models before passing it through Dense Layers to get the output. This hybrid model provides a higher level of accuracy and outperforms LSTM only model.
Outcomes from the Americas World Summit on Artificial Intelligence
I really enjoyed my participating in the World Summit even though it was held virtually on an online platform. We need to learn to adapt to the changing world situation and outbreak of the pandemic and not let it stop our collective education, learning, and networking.
I was able to record and deliver a very good presentation on the topic of machine learning and exoplanetary data. It was very well received. I received a lot of feedback and applause for my presentation and was able to connect with many people. I was also happy to note that my presentation turned out to be the most-watched talk of the Summit.
I also enjoyed listening to many of the other presenters. I learned a lot about how deep learning model works watching the presentation of Google Researcher Samy Bengio. Prof. Greg Dudek from McGill University, talk on using machine learning to deliver autonomy in robots was very informative. Alexis Hannart from Axionable, a sustainable AI services company talked about using Artificial Intelligence for climate change, and Audrey Boguchwal from Samasource raised the issue of biases in AI in her presentation.
I also very much enjoyed listening to the panel discussion on Governing AI: putting principles into practice that included panel members: Jana Novohrdaska, AI Consultant to Deputy Prime Minister’s Office of Slovakia, Mina J Hanna Co-Chair IEEE-SA Global Initiative on Ethics of AI Systems Policy Committee and Yannick Meneceur Policy Advisor on Digital Transformation and Artificial Intelligence at The Council of Europe.
I look forward to speaking and participating in the future editions of the World Summit on Artificial Intelligence.
Winners: Micro:bit Challenge North America Runners Up 2020. NASA SpaceApps 2019, 2018, 2017, 2014. Imagining the Skies 2019. Jesse Ketchum Astronomy Award 2018. Hon. Mention at 2019 NASA Planetary Defense Conference. Emerald Code Grand Prize 2018. Canadian Space Apps 2017.