Technical Reports:

(4)  A. Adler and A. Painsky

      "Feature Importance in Gradient Boosting Trees with Cross-Validation Feature Selection",

      Submitted to the Expert Systems with Application, Under Review,  Sep 2021

      Online version - [link]

(3)  Y. Shalev A. Painsky and I. Ben-Gal, 

      "Neural Joint Entropy Estimation",

      Submitted to the IEEE Transactions on Neural Networks and Learning Systems, Under Review,  Mar 2021

      Online version - [link]

(2)  S. Rosset, R. Heller, A. Painsky and E. Aharoni, 

      "Optimal Procedures for Multiple Testing Problems",

      Submitted to the Annals of Statistics, Under Review,  Nov 2018

      Online version - [link]

(1)  R. Shwartz-Ziv*, A. Painsky* and N. Tishby, 

      "Representation Compression and Generalization in Deep Neural Networks",

      Technical Report, Jul 2018

*Authors contributed equally

Journal Papers:

(12)  A. Painsky and M. Feder 

       "Robust Universal Inference",

        Entropy, Special Issue on Application of Information Theory in Statistics,

        Vol 23, Issue 6, Jun 2021 [link]

(11)  A. Painsky, M. Feder and N. Tishby, 

       "Non-linear Canonical Correlation Analysis: A Compressed Representation Approach",

        Entropy, Special Issue on Theory and Applications of Information Theoretic Machine Learning,

        Vol 22, Issue, 2, Feb 2020 [link]

(10)  A. Painsky and G. W. Wornell, 

       "Bregman Divergence Bounds and Universality Properties of the Logarithmic Loss",

        IEEE Transactions on Information Theory, Vol, 66, Issue 3, Mar 2020 [link]

(9)  A. Painsky, S. Rosset and M. Feder, 

      "Innovation Representation with Application to Causal Inference",

       IEEE Transactions on Information Theory, Vol. 66, Issue 2, Feb 2020 [link]

 

(8)  A. Painsky and S. Rosset, 

      "Lossless Compression of Random Forests",

      Journal of Computer Science and Technology, Vol. 34, Issue 2, Mar 2019 [link]

 

(7)  A. Painsky, S. Rosset and M. Feder, 

      "Linear Independent Component Analysis over Finite Fields: Algorithms and Bounds",

      IEEE Transactions on Signal Processing, Vol. 66, Issue 22, Nov 2018 [link]

 

(6)  A. Painsky and N. Tishby,  

      "Gaussian Lower Bound for the Information Bottleneck Limit",

      Journal of Machine Learning Research (JMLR), Vol. 18, Issue 1,  Apr 2018 [link]

 

(5)  A. Painsky, S. Rosset and M. Feder,

       "Large Alphabet Source Coding using Independent Component Analysis",

       IEEE Transactions on Information Theory, Vol. 63, Issue 10,  Oct 2017 [link]

 

(4)  A. Painsky and S. Rosset, 

       "Cross-Validated Variable Selection in Tree-Based Methods Improves Predictive Performance",

       IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), Vol. 39, Issue 11, Dec 2016 [link]

 

(3)  A. Painsky, S. Rosset and M. Feder, 

       "Generalized Independent Component Analysis over Finite Alphabets",

       IEEE Transactions on Information Theory, Vol. 62, Issue 2, Feb 2016 [link]

 

(2)  A. Painsky and S. Rosset, 

       "Isotonic Modeling with Non-differentiable Loss Functions with Application to Lasso Regularization",

       IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), Vol. 38, Issue 2, Feb 2016 [link]

 

(1)  A. Painsky and S. Rosset, 

       "Optimal Set Cover Formulation for Exclusive Row Biclustering of Gene Expression",

       Journal of Computer Science and Technology, Vol. 29, Issue 3, Apr 2013 [link]

Competitive Conference papers (less than 10% acceptance rate):

 

(2)  A. Painsky and S. Rosset, 

      "Compressing Random Forests",

      IEEE 16th International Conference on Data Mining (ICDM), Dec 2016 [link]

 

(1)  A. Painsky and S. Rosset, 

      "Exclusive Row Biclustering for Gene Expression Using a Combinatorial Auction Approach",

      IEEE 12th International Conference on Data Mining (ICDM), Dec 2012 [link]

 

Conference papers:

(7)  A. Painsky

      "Refined Convergence Rates of the Good-Turing Estimator",

       IEEE Information Theory Workshop (ITW), to appear

(7)  A. Painsky and G.W. Wornell, 

      "On the Universality of the Logistic Loss Function",

       IEEE International Symposium on Information Theory (ISIT), May 2018 [link]

(6)  A. Painsky, S. Rosset and M. Feder, 

      "Binary Independent Component Analysis: Theory, Bounds And Algorithms",

      IEEE 26th International Workshop on Machine Learning for Signal Processing (MLSP), Sep 2016 [link]

 

(5)  A. Painsky, S. Rosset and M. Feder, 

      "A Simple and Efficient Approach for Adaptive Entropy Coding Over Large Alphabets",

       Data Compression Conference (DCC), Apr 2016 [link]

 

(4)  A. Painsky, S. Rosset and M.Feder,  

      "Universal Compression of Memoryless Sources over Large Alphabets via Independent Component Analysis",

       Data Compression Conference (DCC), Apr 2015 [link]

 

(3)  A. Painsky, S. Rosset and M. Feder, 

       "Generalized Binary Independent Component Analysis",

       IEEE International Symposium on Information Theory (ISIT), Jul 2014 [link]

 

(2)  A. Painsky, S. Rosset and M. Feder, 

      "Memoryless Representation of Markov Processes",

       IEEE International Symposium on Information Theory (ISIT), Jul 2013 [link]

 

(1)  A. Painsky

      "First Order Multiple Hypothesis Tracking for the Global Nearest Neighbor Correlation Approach",

      IEEE Workshop on Sensor Data Fusion, Sep 2010. [link] 

 

 

 

PhD thesis (Statistics):

Generalized Independent Component Analysis over Finite Alphabets, Tel Aviv University, 2016 

Online version [link]

 

Matser thesis (Statistics):

Exclusive Row Biclustering for Gene Expression Using a Combinatorial Auction Approach, Tel Aviv University, 2011 

Online version [link]

 

Book Chapters:

(1)  A. Painsky

      "Quality Assessment and Evaluation Criteria in Supervised Learning",

       The Handbook of Machine Learning for Data Science, Springer Publishing. To Appear Mar 2020