My current interests lie broadly in the intersection of perception, computer vision and deep learning. I work with Prof. Marios Savvides, at the CyLab Biometrics Center. I also occasionally collaborate with Prof. Ole Menshoel. Some of my work was inspired by the work of and discussions with Prof. Tomaso Poggio.

Much of my work is centered around the problem of perception and representation learning. Building useful representations of data is of paramount importance. Ideally, a single good representation would be useful for any task one is interested in. Towards this goal, representations that are invariant to nuisance within-class transformations and that are discriminative towards between class transformations are useful. Some of my work has explored learning such representations by building invariant kernel classifiers, loss functions that promote such invariance and finally inductive biases in network architectures that explicitly promote invariance during feedforward computations. Self-supervision has also emerged as a promising class of techniques that promote rich representation learning through pretext tasks. Some of my work aims to provide a better understanding of the phenomenon of self-supervision paving the way for the development of more effective techniques. Finally, a few of my studies have dealt with inverse problems from the deep learning and randomized sparse signal approximation perspective.