Skip to content. | Skip to navigation

You are here: Home » CSCI » Pressroom » News » Alumni Interview Series: Ajani Stewart – Postbaccalaureate Researcher, MIT
Inside
 
Document Actions

Alumni Interview Series: Ajani Stewart – Postbaccalaureate Researcher, MIT

To offer some perspective on the achievements of our department's students, we're going to be posting a series of short interviews with some of its distinguished alumni. First up in this series is Ajani Stewart! Ajani graduated from Hunter in 2021 and is now engaged in postbaccalaureate research. Read on for more details from Ajani himself in the following Q&A.

Thanks for taking the time to speak with us! What are you working on these days?

Right now I’m at MIT working as a post-bacculaurate research scholar in the Brain and Cognitive Sciences Department. The lab I’m in, the McDermott Lab, studies the computational nature of human hearing. My project is studying how humans localize multiple sounds at once and how attention and working memory affect that. I do this by using neural networks created in the lab and designed to localize one sound, and extend that model to be able to localize multiple sounds at once. I then compare these networks to humans doing a similar multisource localization task.

Thinking in terms of your current projects - school or work - can you tell us a little about how studying at Hunter helped you get there?

Machine learning is at the heart of the research I am doing. Most of the research in my lab, and a significant amount of research in the department, uses machine learning models in some way. Learning about machine learning at Hunter from Dr. Susan Epstein has prepared me significantly for the research I do and see everyday. Further, learning about machine learning also introduced me to the exciting world of research and I would not be where I am now without it.

What do you find exciting about your field right now?

There is a growing trend to use neural networks as models of sensory perception in humans. This is exciting because there is research showing that neural networks optimized to perform tasks like sound localizations or object recognition can exhibit behaviors similar to humans and have internal representations that are similar as well. If we have good models of sensory systems that can make accurate predictions about those systems, we can use them to glean insights into how the human systems actually work. For example, we could accurately simulate the effects of damage to the retina and figure out what interventions we could do to mitigate or reverse that damage. This could aid the development of neural prosthesis and possibly restore sight to blind people.

What do you see yourself doing downline? Say, 5 years from now: will you be working more deeply in the area you are now, or branching out to something else?

I’m not really sure what I will be doing in the future. I have just finished applying to graduate school so in 5 years I will be likely finishing be finishing that up. I am always open to new and interesting problems so I would not be surprised if I completely switch fields.

Serious stuff aside, what do you miss about your time at Hunter?

I miss the Hunter environment the most. There is always something going and there was never a dull day. I also miss the halal carts near hunter.

Is there an album or artist that you like to listen to when you need to code, or some other tool/practice that helps you get in the zone?

I usually listen to math rock/progressive rock these days when I code and my go to album is “Remember That You Will Die” by Polyphia. I also like listening to heavy metal and my go to album for that is “Snake Behind the Sun” by Shy, Low.