Andrew E Gelfand

Research Scientist in Machine Learning / Quantitative Researcher

I am an applied researcher interested in developing and deploying scalable machine learning systems. I work at a NYC-based hedge fund where I apply advanced machine learning techniques to complex, quantitative trading problems. Prior to that I was a research scientist at Yahoo Labs working on recommendation and ranking problems.

I recieved my Ph.D. from the University of California, Irvine and was co-advised by Rina Dechter and Alex Ihler. I also had the good fortune of working with Max Welling and Misha Chertkov during my time at UC Irvine. My expertise is in developing methods to efficiently learn from data using graphical models - a modeling formalism that provides structure to probability distributions over large, complex systems.

Publications

Geographic Segmentation via a Latent Poisson Factor Model
Rose Yu, Andrew Gelfand, Suju Rajan, Cyrus Shahabi & Yan Liu
WSDM 2016. San Francisco, USA, February 2016.
A Graphical Transformation for Belief Propagation: Maximum Weight Matchings and Odd-Sized Cycles
Jinwoo Shin, Andrew E. Gelfand and Misha Chertkov
NIPS 2013. Lake Tahoe, NV, USA, December 2013.
Loop Calculus and Bootstrap-Belief Propagation for Perfect Matchings on Arbitrary Graphs
Misha Chertkov, Andrew E. Gelfand and Jinwoo Shin
In Proceedings of the International Meeting on 'Inference, Computation, and Spin Glasses'. Sapporo, Japan, July 2013.
Belief Propagation for Linear Programming
Andrew E. Gelfand, Jinwoo Shin and Misha Chertkov
ISIT 2013. Istanbul, Turkey, July 2013.
Generalized Belief Propagation on Tree Robust Structured Region Graphs    [Proofs]
  Andrew E. Gelfand and Max Welling
  UAI 2012. Catalina Island, CA, USA, August 2012.
A Cluster-Cumulant Expansion at the Fixed Points of Belief Propagation
Max Welling, Andrew E. Gelfand and Alexander Ihler
UAI 2012. Catalina Island, CA, USA, August 2012.
Integrating Local Classifiers through Nonlinear Dynamics on Label Graphs with an Application to Image Segmentation
Yutian Chen, Andrew Gelfand, Charless Fowlkes & Max Welling
ICCV 2011. Barcelona, Spain, November 2011.
Stopping Rules for Randomized Greedy Triangulation Schemes
Andrew E. Gelfand, Kalev Kask and Rina Dechter
AAAI 2011. San Francisco, CA, USA, August 2011.
Pushing the Power of Stochastic Greedy Ordering Schemes for Inference in Graphical Models
Kalev Kask, Andrew E. Gelfand, Lars Otten, and Rina Dechter
AAAI 2011. San Francisco, CA, USA, August 2011.
On Herding and the Perceptron Cycling Theorem
Andrew Gelfand, Laurens van der Maaten, Yutian Chen & Max Welling. NIPS 2010. Vancouver, Canada, December 2010.
BEEM : Bucket Elimination with External Memory
Kalev Kask, Rina Dechter and Andrew E. Gelfand.
UAI 2010. Catalina Island, CA, USA, July 2010.

Book Chapters

Herding for Structured Prediction
  Yutian Chen, Andrew E. Gelfand and Max Welling.
  In Advanced Structured Prediction, MIT Press 2014.
  S. Nowozin, P. Gehler, J. Jancsary, C. Lampert (editors).