Noam Shental

 MyLastLame @ openu.ac.il
  Phone: 972-9-7781252 
 

 noam shental photo

 About Me

On October 2008 I joined the CS department at the Open University of Israel, as a faculty member. Previously, I was a postdoctoral fellow at the Department of Physics of Complex System at the Weizmann Institute of Science, Rehovot, Israel, working with Prof. Eytan Domany. I received my Ph.D. in Computational Neuroscience, from the Hebrew University, supervised by Prof. Daphna Weinshall.
Brief CV.

potential master students at the Open U are most welcome to email me

 
כתבות בעתונות הכללית
 אסף שטול טראורינג, "שיטה ישראלית תוזיל בדיקות גנטיות לגילוי מחלות" עיתון הארץ, 11.2.2011 קישור להארץ
צבי צורי, "מוטציות בערימה של גנים", "גליליאו" 150, פברואר 2011 

Research Interests

  • Application of Compressed Sensing in biology.
  • Applications of Machine Learning algorithms to real life problems.
  • Algorithms for approximate inference and their applications to real-world problems.

Projects

  • A Compressed Sensing application in Genetics -  A novel application of Compressed Sensing to the problem of indentifying novel mutations and their carriers in large cohorts of DNA samples via Next Generation Sequencing technology. The same method is also able to locate carriers of known mutations, e.g., in the case of genetic screening (pdf). For a general overview of the method, see the above write ups in Israeli newspapers (in Hebrew).   
  • The antigen microarray - A high throughput device which measures an "immunological profile" of a person based on a blood sample. I was involved in the algorithmic part of this project, which heavily resides on machine learning techniques. This was my main postdoctoral project, and later I headed the bioinformatics division in a startup company, ImmunArray Inc., Rehovot, Israel, which was established in order to pursue the project. 
  • Algorithms for semi-supervised learning - We dealt with two scenarios in semi-supervised learning. Firstly, the .classical. scenario of a large unlabelled data set which is accompanied by a small labelled set (pdf). Secondly, we considered the scenario where partial supervision is provided in the form of equivalence constraints (pdf), which can also be considered as a constrained clustering problem (pdf).
  • Applications of graphical models for clustering and segmentation . We represented the problem of data clustering as an inference problem in an undirected graphical model, and applied Generalized Belief Propagation in order to solve it (pdf).
  • A Graphical Models approach for a storage (hard disks) and communications (cellular phones) applications  - We mapped specific hard problems in the field of electrical engineering to the field of Graphical models, and solved them almost optimally (pdf).

 

Publications

Journal Publications

  • Identification of rare alleles and their carriers using compressed se(que)nsing
Noam Shental^, Amnon Amir and Or Zuk.
Nucleic Acids Research, 38(19):e179, 1-22, Aug 2010
^ Corresponding author
ComSeq code (matlab)

  • An antibody core signature of Systemic Lupus Erythematosus detected by antigen microarray
Ittai Fattal*, Noam Shental*, Dror Mevorach , Juan-Manuel Anaya, Avi Livneh, Pnina Langevitz, Gisele Zandman-Goddard, Rachel Pauzner, Miriam Lerner, Miri Blank, Maria-Eugenia Hincapie, Uzi Gafter, Yaakov Naparstek, Yehuda Shoenfeld, Eytan Domany and Irun R. Cohen.
Immunology (130): 337-343, 2009.
* Equal contribution
   
  • Discrete-Input Two-Dimensional Gaussian Channels With Memory: Estimation and Information Rates via Graphical Models and Statistical Mechanics.
    Ori Shental*, Noam Shental*, Shlomo Shamai (Shitz), Ido Kanter, Anthony Weiss, and Yair Weiss.
    IEEE Transactions on Information Theory (TIT), 54(April): 1500-1513, 2008.
    * Equal contribution
Awarded Best Student Paper of the Year 2008 in Signal Processing and Coding for Data Storage, by the IEEE Communications Society Data Storage Technical Committee.

 

  • Learning a Mahalanobis Metric from Equivalence Constraints.
    Aharon Bar-Hillel, Tomer Hertz, Noam Shental and Daphna Weinshall.
    Journal of Machine Learning Research (JMLR) 6(Jun): 937-965, 2005. More than 150 citations

Book Chapters

  • Gaussian Mixture Models with Equivalence Constraints.  
    Noam Shental, Aharon Bar-Hillel, Tomer Hertz and Daphna Weinshall.
    in "Constrained Clustering: Advances in Algorithms, Theory, and Applications", edited by S. Basu, I. Davidson, and K.L. Wagstaff, Chapman & Hall/CRC Data Mining Series, 2008. (for example in alibris.com)

 

Peer-Reviewed Conference Publications

  • Semi-Supervised Learning - A Statistical Physics Approach.
    Gad Getz*, Noam Shental* and Eytan Domany.
    Workshop on "Learning with Partially Classified Training Data", 22nd International Conference on Machine Learning (ICML), 2005.
    * Equal contribution
  • On the Achievable Information Rates of Finite-State Input Two-Dimensional Channels with Memory.
    Ori Shental, Noam Shental and Shlomo Shamai (Shitz).
    International Symposium For Information Theory (ISIT), 2005.
  • Generalized Belief Propagation Receiver for Near-Optimal Detection of Two-Dimensional Channels with Memory.
    Ori Shental, Noam Shental, Anthony Weiss and Yair Weiss.
    IEEE Information Theory Workshop (ITW), 2004.
  • Computing Gaussian Mixture Models with EM using Equivalence Constraints.
    Noam Shental, Aharon Bar-Hillel, Tomer Hertz and Daphna Weinshall.
    Neural Information Processing Systems (NIPS), 2003. More than 100 citations
  • Pairwise Clustering and Graphical Models.
    Noam Shental, A. Zomet, Tomer Hertz and Yair Weiss.
    Neural Information Processing Systems (NIPS), 2003.
  • Learning and Inferring Image Segmentations using the GBP Typical Cut.
    Noam Shental, Assaf Zomet, Tomer Hertz and Yair Weiss.
    9th International Conference on Computer Vision (ICCV), 2003.
  • Computing Gaussian Mixture Models with EM using Side Information.
    Noam Shental, Aharon Bar-Hillel, Tomer Hertz and Daphna Weinshall.
    Workshop on "The Continuum from Labeled to Unlabeled Data in Machine Learning and Data Mining",
    20th International Conference on Machine Learning (ICML), 2003.
  • Learning Distance Functions using Equivalence Relations.
    Aharon Bar-Hillel, Tomer Hertz, Noam Shental and Daphna Weinshall.
    20th International Conference on Machine Learning (ICML), 2003. More than 200 citations
  • Enhancing Image and Video Retrieval: Learning via Equivalence Constraints.
    Tomer Hertz, Aharon Bar-Hillel, Noam Shental and Daphna Weinshall.
    IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2003.
  • Adjustment Learning and Relevant Component Analysis.
    Noam Shental, Tomer Hertz, Daphna Weinshall and M. Pavel.
    7th European conference of Computer Vision (ECCV), 2002. More than 90 citations
  • Perceptual Grouping and Segmentation by Stochastic Clustering.
    Yoram Gdalyahu, Noam Shental and Daphna Weinshall.
    Conference on Computer Vision and Pattern Recognition (CVPR), 2000.

Dissertation

  • From Unsupervised to Semi-Supervised Learning: Algorithms and Applications
    PhD. thesis, Hebrew University, 2004.

Code

  • Compressed Sequencing (ComSeq)  – matlab implementation
  • Relevant Component Analysis (RCA)  - matlab implementation
  • Constrained EM algorithm  - stand-alone matlab implementation which includes the BNT package.
  • GBP for image segmentation - matlab implementation.

 

Last updated 23.2.11