OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks
Learning Factored Representations in a Deep Mixture of Experts
Understanding Deep Architectures using a Recursive Convolutional Network
Previously, I was in San Francisco working on antispam and antimalware at Cisco IronPort Systems.
Before IronPort, I was at NetApp, where I worked on the WAFL filesystem and rewrote much of the NVLog intent journal. Before that, I was at Brown, where I studied math, computer science and cognitive science.
At Brown, I developed a set of visualization tools for differential geometry with Prof. Banchoff. This has since been applied to more applications and labs over the past few years; one of the latest complements the textbook "Differential Geometry of Curves and Surfaces" by Banchoff and Lovett.
For Burning Man '09, I made a large jigsaw puzzle.
Projects older than those on my CV are here. This includes selected final projects from various classes.