Alejandro (Alex) was born and raised in Mexico City, Mexico, until the age of 7 when he moved to Oaxaca, Mexico. At the age of 18, he moved to Toronto to start his studies at the University of Toronto. He is currently heading onto his last year as an undergraduate student with a double major in Mathematics and Physics & Astrophysics. Outside of the classroom, Alex loves to read books in his free time. Now reading Stoic Foundations, The 48 Laws of Power and a book in Mathematical Analysis. Alex enjoys to run/bike/hike after SURP.
What made you decide to participate in SURP?
Last summer I worked on machine learning models to power astrophysical research with Alexander Laroche and Professor J. Speagle. That experience sparked a strong interest in me in how machine learning works, and how we can use its inner workings to get the best results when applying it. Over the past year, I have continued doing research with them, exploring more complex ideas in machine learning to foster scientific progress. Thanks to this experience, my SURP supervisors were happy for me to collaborate more independently in the project and participate as a more mature researcher. That is one of my favorite parts: the freedom and support I have had to grow as a researcher. Despite my age, I can contribute ideas and shape the project. It is an incredible experience as I prepare for a PhD.
What is your favourite thing about SURP?
All of my supervisors are astrophysicists, but they all have different focuses. And this perfectly matches my project where I use advanced machine learning techniques to try to answer astrophysical questions. It is a mix of both worlds! Then, during a general meeting I present my ideas and thoughts to guide the project to my team of four supervisors. As mentioned, each one of them is a worldwide expert on a particular side of the project. From machine learning to the properties of the data. Then, suddenly me, an undergraduate, must convince them all that my ideas and decisions to guide the project are the right ones! My favorite part is that only through a lot of work and preparation, I achieve that constantly. It is challenging and incredible.
Can you tell us about your research project?
We work towards understanding polarisation images in a much better way through machine learning. The idea behind our project relates to how we as humans perceive things. We have five senses, and because of that we can arguably perform better at surviving than if we had only one. For example, our ancestors could sense the attack of wolves because they either saw them or heard them. We understand things better when we have different views of them. Recent advances in machine learning have shown that machine learning models behave in a similar way. More ways of looking at the same data means better results. Then, we work with polarisation data but on different streamlines. By combining the view of stokes I, rotation measure and peak intensity images, we hope to get a much better sense of what an individual polarisation source means. You can think of it as if the machine learning model learns to fusion all these views (images) of what a polarisation source means, onto one concept. That is exactly what we want and what we do. Once we have that, it becomes much easier to study the characteristic features of these “concepts”, which is our end goal. Finally, one key feature of our work is that we work with images. To achieve our goals, we compress images onto a discrete regime. In other words, we transform images into something like “words” which are finite and much easier to study by the latest machine learning approaches. It is not trivial, so if you are more curious about it, email me. It is such fascinating work.
Can you explain how SURP has been different from your undergrad work?
The big main difference of SURP with respect to undergraduate work is the amount of time you have to work a project. And this is great given SURP is a full-time job. In my case, it is quite exciting for me because everyday I have the opportunity to read a new paper, propose a new idea and code it, and perform experiments to validate my past ideas. The way is yet long to go, but I do sometimes feel like an actual researcher as my supervisors. And I like that, because it is the small steps through the daily work which finally give shape to the last one. SURP gives me the perfect opportunity to keep on paving my way.
What are your plans for the future?
My short and long life goals align with my passion for machine learning. I am eager to understand formally how these systems work. They are complex, and I find beauty in that. It is awesome how through mathematics we can give the sense of “person” or “human response” to a computer. Without meaning to offend all the computers out there! As a result, in the next year I see myself joining my master’s degree in mathematics before moving onto a PhD in Machine Learning Theory. I look forward to reading more books and becoming a better person.
Tell us something fun about yourself unrelated to SURP!
- I was shortlisted for the Mexican team in the International Olympiad in Informatics. Like the IMO but for CS!
- I will probably talk about machine learning a huge amount of time.
- I have traveled to two conferences by myself already. One was in Arizona! Both have been amazing.
- I was an Aurora Borealis fellow last year at SURP.
- My supervisor Dr. Speagle has talked about my research at Harvard during the Astro AI Workshop 2024 and at the CCA at the Flatiron Institute.
The Dunlap Institute for Astronomy and Astrophysics at the University of Toronto is an endowed research institute with over 80 faculty, postdocs, students, and staff, dedicated to innovative technology, ground-breaking research, world-class training, and public engagement.
The research themes of its faculty and Dunlap Fellows span the Universe and include: optical, infrared and radio instrumentation, Dark Energy, large-scale structure, the Cosmic Microwave Background, the interstellar medium, galaxy evolution, cosmic magnetism, and time-domain science.
The Dunlap Institute, the David A. Dunlap of Astronomy and Astrophysics, and other researchers across the University of Toronto’s three campuses together comprise the leading concentration of astronomers in Canada, at the leading research university in the country.
The Dunlap Institute is committed to making its science, training, and public outreach activities productive and enjoyable for everyone of all backgrounds and identities.