Many recent advancements in machine learning have come by way of neuroscience: indeed, multilayer perceptrons, convolutional neural networks, and residual blocks each aim to mimic many concepts gleaned from our understanding of the brain and fuse them with the traditional statistical methods of AI. Machine learning has similarly impacted neuroscience, where an increased emphasis on studying larger network structure and function has been made possible with certain machine learning and computer vision methods. Matthew is interested in performing research at the intersection of these two fields, where continuing to search for deeper insights is certain to be of benefit to both.

With Bachelor’s of Science degrees in computer science and computer engineering and additional minors of mathematics and physics, Matthew is currently pursuing a master’s in computer science. His current work in computer vision includes domain adaptation, fine-grained visual classification on long-tailed “in-the-wild” data, and instance segmentation of 2D and 3D cellular data, the latter of which will be used to create a tool for biology researchers for automatically counting cells.

Outside of his work, Matthew enjoys rock climbing, trail running, cooking, and reading and writing about topics related to machine learning, cognitive neuroscience, and the ethics involved with studying and developing technology for both.