Volodymyr Kuleshov

Ph.D. Candidate
Deparment of Computer Science
Stanford University


My thesis work focuses on machine learning and its applications in genomics and personalized medicine. I am intereted in:

  • Fast approximate inference in probabilistic models, particularly recent methods based on deep learning ICLR16
  • Uncertainty estimation techniques, e.g. calibrated probabilistic forecasting NIPS15a NIPS15b

On the applications side, my interests/projects include:

  • Machine reading of the scientific literature with the goal of making medical knowledge easily accessible to practitioners
  • Identifying the genetic causes of disease via new mathematical models and inference algorithms that can scale to very large patient cohorts
  • New genome sequencing technologies that combine existing wetlab techniques with new statistical methods, thus making them significantly more affordable and accurate Nat. Biotech. 14

Previously, I worked on computational methods in economics and game theory, particularly on designing efficient markets. At Stanford, my collaborators include Serafim Batzoglou, Michael Snyder, Percy Liang, and Stefano Ermon.

In 2012-2013, I spent a year off at Moleculo, where I developed algorithms that now power Illumina's human phasing service.

Also, check out also my online introductory course notes on probabilistic graphical modeling, and follow me on my new Twitter account!

Papers


Machine learning

Neural variational random field learning.
Volodymyr Kuleshov and Stefano Ermon.
International Conference on Learning Representations (Workshop track), 2016


Reliable confidence estimation via online learning.
Volodymyr Kuleshov and Stefano Ermon.
Machine Learning for Healthcare workshop at NIPS, 2015


Calibrated structured prediction.
Volodymyr Kuleshov and Percy Liang.
Neural Information Processing Systems, 2015


Scaling up simultaneous matrix diagonalization.
Volodymyr Kuleshov*, Arun Chaganty*, Percy Liang.
Optimization for Machine Learning Workshop at NIPS, 2015


Tensor factorization via matrix factorization.
Volodymyr Kuleshov*, Arun Chaganty*, Percy Liang.
Artificial Intelligence and Statistics, 2015


Fast algorithms for sparse principal component analysis based on Rayleigh quotient iteration.
Volodymyr Kuleshov.
International Conference on Machine Learning, 2013


Algorithms for multi-armed bandit problems.
Volodymyr Kuleshov and Doina Precup.
Manuscript



Genomics

High-resolution structure of the human microbiome revealed with synthetic long reads.
Volodymyr Kuleshov, Chao Jiang, Wenyu Zhou, Fereshteh Jahanbani, Serafim Batzoglou, Michael Snyder.
Nature Biotechnology, 2015 (Advance Online Publication)


Synthetic long read technologies in genome phasing and beyond.
Volodymyr Kuleshov, Dan Xie, Rui Chen, Dmitry Pushkarev, et al.
Intelligent Systems for Molecular Biology (Highlights Track), 2015.


Probabilistic single-individual haplotyping.
Volodymyr Kuleshov.
European Conference on Computational Biology, 2014.


Whole-genome haplotyping using long reads and statistical methods.
Volodymyr Kuleshov, Dan Xie, Rui Chen, Dmitry Pushkarev, et al.
Nature Biotechnology, 2014



Algorithmic game theory

Inverse game theory: learning utilities in succinct games.
Volodymyr Kuleshov and Okke Schrijvers.
Web and Internet Economics, 2015
World Congress of the Game Theory Society (Contributed Talk), 2016


On the efficiency of the simplest market mechanisms.
Volodymyr Kuleshov and Gordon Wilfong.
Web and Internet Economics, 2012


On the efficiency of markets with two-sided proportional allocation mechanisms.
Volodymyr Kuleshov and Adrian Vetta.
Algorithmic Game Theory, 2010


Contact


Volodymyr Kuleshov
Clark Center, Room S260
318 Campus Drive
Stanford, CA 94305
E: [last name]@stanford.edu
map generatorhttp://www.stromvergleich-uebersicht.de