CS 4803DL / 7643 (Deep Learning) see here for course website


I am an Assistant Professor at the School of Interactive Computing in the College of Computing. I am also affiliated with the Georgia Tech Research Institute and serve as an Associate Director of ML@GT which is the machine learning center recently created at Georgia Tech. Previously I was a Research Scientist at SRI International Sarnoff in Princeton, and before that received my Ph.D. in 2010 with Professor Ron Arkin as my advisor.

My areas of research specifically focus on the intersection of learning methods for sensor processing and robotics, developing novel machine learning algorithms and formulations towards solving some of the more difficult perception problems in these areas. I am especially interested recently in unsupervised/semi-supervised methods, continual/lifelong learning, and distributed perception.



Three ICLR accepted papers on learning without explicit class labels for supervised, cross-task, and semi-supervised learning as well as goal-driven navigation (with Salesforce) and few-shot learning (with VT). Code for the first two are out here and here, and forthcoming for the last one.


Pytorch benchmark suite and framework for continual/lifelong learning released! [ paper | code ]


Accepted NIPS Continual Learning Workshop paper with a categorization of current scenarios and strong baselines achieving state of art (or close) on current continual learning scenarios!


Accepted WACV paper on PCB board component detection with graph networks [arxiv]


Teaching Deep Learning cross-listed undergraduate and graduate. Website here.


New ONR funding for distributed perception and planning with Lu Feng (UVA), Pratap Tokekar (VT), and Ufuk Topcu (UT Austin).


Journal paper on TS-LSTMs accepted


Pytorch code release for our clustering work on github


CVPR Deep-Vision Workshop paper on a new probabilistic deep learning clustering formulation [arxiv]


News article about new DARPA lifelong learning project


Accepted CVPR paper: Higher-Order Object Interactions for Video Understanding [arxiv]