Robert Preston Hess
MIT EECS | Philips Undergraduate Research and Innovation Scholar
Modeling Auditory Attention with Machine Learning
- Brain and Cognitive Science
Attentional selection allows humans to recognize communication signals amid concurrent sound sources (the cocktail party problem”). Although attentional abilities have been characterized to some extent in humans, we lack quantitative models that can account for attention-mediated behavior, explain the conditions in which attentional selection should succeed or fail, and reveal how attention should influence neural representations to enable selective listening. Inspired by neurophysiological observations of attention, we will develop a model of auditory attention by equipping a neural network with stimulus-dependent gains. We will optimize the model to perform a recognition task on spatialized audio signals, reporting the words or the location of a cued talker in a multi-source mixture.”
I will use my time in the SuperUROP program to dive more deeply into the research that I have pursued at MIT. I hope to use the time in the program to prepare myself for graduate school through presentations, posters, and hopefully a publication at the end of my time. The project will allow me to expand my understanding of both cognitive and computer science as I apply for programs in their intersection.