Daniel (Dan) Sedra

Applied Scientist
BingAds @ Microsoft
Bellevue, WA, 94003
17 Bravern II
dmsedra at gmail dot com
(Resume, CV, LinkedIn)

 

Bio:

I was an applied scientist in the BingAds group @ Microsoft. I worked on diverse problems including click prediction, causality modeling, feature selection, intent inference, and ranking.

Before that, I got my MSc in the Computer Science Department at Cornell University under Professor Kilian Weinberger. My research mainly focused on deep learning for image classification. I worked on various stochastic training paradigms for deep residual networks, including: stochastic depth (see [1] below), stochastic batch normalization, and network expansion. I also worked on collaborative crowdsourcing (see [2] below) for image annotation.

I received my BSc in Computer Science from Carnegie Mellon University in 2014. My undergraduate thesis advisor was Professor William Cohen.


Work Experience:


Publications:

  1. Huang, G., Sun, Y., Liu, Z., Sedra, D. , and Weinberger, K. Deep Networks With Stochastic Depth. In the European Conference on Computer Vision , 2016. (pdf, arxiv)
  2. Upchurch, P., Sedra, D. , Mullen, A., Bala, K., Hirsh, H. Interactive Consensus Agreement Games for Image Labeling. In the Fourth AAAI Conference on Human Computation and Crowdsourcing, 2016. (pdf)


Projects:


Teaching Experience: