Curriculum Vitae



Personal data
Family name Szörényi
Given name Balázs
Date of birth May 6, 1976
Place of birth Szeged, Hungary
Citizenship Hungarian
Office address H-6701 Szeged, Postafiók 652., Hungary
E-mail
click here to reveal
Language skills Hungarian: native
English: fluent
German: good
French: intermediate (between A2 and B1)


Positions held
2015 - Postdoc at the Technion, Haifa, Israel
2008 - Associate research fellow at the Research Group on Artificial Intelligence of the Hungarian Academy of Sciences, University of Szeged (RGAI)
2013 - 2014 Postdoc at INRIA Lille-Nord Europe (SequeL group) for 2 years
2008 - 2009 Postdoc (Wissenschaftliche Mitarbeiter) at the Ruhr-Universität Bochum for 1 year
2003 - 2008 Funded research assistant at the Research Group on Artificial Intelligence of the Hungarian Academy of Sciences, University of Szeged (RGAI)


Education
2008 Ph.D. degree at the University of Szeged.
Thesis: Learnability and Characterization Results for Classes of Boolean Functions.
Supervisor: György Turán
2004
M.Sc. degree in Mathematics at the University of Szeged
2000
M.Sc. degree in Computer Science/Economics at the University of Szeged
1997
B.Sc. degree in Computer Science at the József Attila University, Szeged


Research Area
Online learning
Learning theory and the theory of machine learning
Theory of Boolean functions


Teaching Experience (at the University of Szeged)
Lectures The theory of machine learning
Approximation algorithms
Introduction to reinforcement learning
Algorithms in bioinformatics

Tutorials
Algorithms and data structures I
Numerical analysis I-II
Operations research II
Introduction to programming
Logic programming
Artificial Intelligence I





Publications



Papers in journals
  • Róbert Busa-Fekete, Balázs Szörényi, Paul Weng, Weiwei Cheng, Eyke Hüllermeier:
    Preference-based Reinforcement Learning: Evolutionary Direct Policy Search using a Preference-based Racing Algorithm
    Machine Learning, 97(3): 327 --351, 2014.
    DOI 10.1007/s10994-014-5458-8
    [download]
  • Malte Darnstädt, Hans Ulrich Simon, and Balázs Szörényi:
    Supervised learning and Co-training,
    Theoretical Computer Science, 519: 68 -- 87, 2014.
    [download preprint, download from Elsevier ]
  • Hans Ulrich Simon, and Balázs Szörényi:
    One-inclusion Hypergraph Density Revisited,
    Information Processing Letters, 110(8-9): 341 -- 344, April 2010.
    [download the latest version]
  • Robert H. Sloan, Balázs Szörényi, and György Turán:
    On k-term DNF with largest number of prime implicants.
    SIAM Journal on Discrete Mathematics, 21(4): 987 -- 998, Jan. 2008.
    [download]
  • Robert H. Sloan, Balázs Szörényi, and György Turán:
    Projective DNF formulae and their revision.
    Discrete Applied Mathematics, 156: 530 -- 544, 2008.
    [download]
  • Robert H. Sloan, Balázs Szörényi, and György Turán:
    Revising threshold functions.
    Theoretical Computer Science, 382(3): 198 -- 208, 2007. (ALT 2004 Special Issue.)
    [download]
  • Balázs Szörényi:
    Disjoint DNF tautologies with conflict bound two.
    Journal on Satisfiability, Boolean Modeling and Computation, 4: 1 -- 14, 2007.
    [download from JSAT]
  • Judy Goldsmith, Robert H. Sloan, Balázs Szörényi, and György Turán:
    Theory revision with queries: Horn, read-once, and parity formulas.
    Artificial Intelligence, 156:139 -- 176, 2004.
    [download]


Papers in conference proceedings
  • Nathan Korda, Balázs Szörényi, Shuai Li:
    Distributed Clustering of Linear Bandits in Peer to Peer Networks.
    33rd International Conference on Machine Learning (ICML 2016),
    [arXiv version]
  • Róbert Busa-Fekete, Balázs Szörényi, Krzysztof Dembczyński, Eyke Hüllermeier:
    Online F-Measure Optimization.
    Twenty-ninth Annual Conference on Neural Information Processing Systems (NIPS 2015),
    595--603, 2015.
    [download]
  • Balázs Szörényi, Róbert Busa-Fekete, Adil Paul, Eyke Hüllermeier:
    Online Rank Elicitation for Plackett-Luce: A Dueling Bandits Approach.
    Twenty-ninth Annual Conference on Neural Information Processing Systems (NIPS 2015),
    604--612, 2015.
    [download]
  • Balázs Szörényi, Róbert Busa-Fekete, Paul Weng, Eyke Hüllermeier:
    Qualitative Multi-Armed Bandits: A Quantile-Based Approach.
    32nd International Conference on Machine Learning (ICML 2015),
    JMLR, W&CP Vol. 37, 1660--1668, 2015.
    [download]
  • Balázs Szörényi, Gunnar Kedenburg, Remi Munos:
    Optimistic Planning in Markov Decision Processes Using a Generative Model.
    Twenty-eighth Annual Conference on Neural Information Processing Systems (NIPS 2014),
    1035--1043, 2014.
    [download]
  • Szabolcs Iván, Ádám D. Lelkes, Judit Nagy-György, Balázs Szörényi, György Turán:
    Biclique coverings, rectifier networks and the cost of epsilon-removal.
    16th International Workshop on Descriptional Complexity of Formal Systems (DCFS 2014)
    2014.
    [arXiv version]
  • Róbert Busa-Fekete, Eyke Hüllermeier, Balázs Szörényi:
    Preference-Based Rank Elicitation using Statistical Models: The Case of Mallows.
    31st International Conference on Machine Learning (ICML 2014),
    JMLR W&CP 32 (1) : 1071--1079, 2014.
    [download from JMLR]
  • Róbert Busa-Fekete, Balázs Szörényi, Eyke Hüllermeier:
    PAC Rank Elicitation through Adaptive Sampling of Noisy Preferences.
    28th AAAI Conference on Artificial Intelligence (AAAI-14),
    AAAI Press, 1701--1707, 2014.
    [download from AAAI]
  • Róbert Busa-Fekete, Balázs Szörényi, Weiwei Cheng, Paul Weng, Eyke Hüllermeier:
    Top-k Selection based on Adaptive Sampling of Noisy Preferences.
    30th International Conference on Machine Learning (ICML 2013),
    JMLR W&CP 28 (3): 1094--1102, 2013
    [download from JMLR]
  • Balázs Szörényi, Róbert Busa-Fekete, István Hegedűs, Róbert Ormándi, Márk Jelasity, Balázs Kégl:
    Gossip-based distributed stochastic bandit algorithms.
    30th International Conference on Machine Learning (ICML 2013),
    JMLR W&CP 28 (3): 19--27, 2013.
    [download from JMLR]
  • Malte Darnstädt, Hans Ulrich Simon, Balázs Szörényi:
    Unlabeled Data Does Provably Help.
    Proceedings of the 30th International Symposium on Theoretical Aspects of Computer Science (STACS 2013).
    Leibniz International Proceedings in Informatics (LIPIcs) 20, 185--196, 2013.
    [download from Dagstuhl homepage]
  • Kira Adaricheva, Robert Sloan, Balázs Szörényi and György Turán:
    Horn Belief Contraction: Remainders, Envelopes and Complexity.
    In Proceedings of the 13th International Conference on Principles of Knowledge Representation and Reasoning, 107 -- 115, 2012.
    In Proceedings of the Tenth International Symposium on Logical Formalizations of Commonsense Reasoning (CS-11), 2011.
    [download from commonsensereasoning.org]
  • Malte Darnstädt, Hans Ulrich Simon, Balázs Szörényi:
    Supervised Learning and Co-training.
    Proceedings of the 22nd International Conference on Algorithmic Learning Theory (ALT 2011).
    Springer LNCS 6925, 425 -- 439, 2011.
    [download]
  • Balázs Szörényi:
    Characterizing Statistical Query Learning: Simplified Notions and Proofs.
    Proceedings of the 20th International Conference on Algorithmic Learning Theory (ALT 2009),
    Springer LNAI 5809, 186 -- 200, 2009.
    [download the extended version]
  • Marina Langlois, Robert H. Sloan, Balázs Szörényi, and György Turán:
    Horn Complements: Towards Horn-to-Horn belief revision.
    Twenty-Third AAAI Conference on Artificial Intelligence (AAAI-08),
    AAAI Press, 466 -- 471, 2008.
    [download]
  • Judy Goldsmith, Robert H. Sloan, Balázs Szörényi, and György Turán:
    Theory revision with queries: results and problems.
    Workshop on Learning with Logic and Logics for Learning (LLLL),
    Japanese Society for AI, 39 -- 44, 2005.
  • Judy Goldsmith, Robert H. Sloan, Balázs Szörényi, and György Turán:
    New revision algorithms.
    15th Algorithmic Learning Theory (ALT 2004), Springer LNAI 3244, 395 -- 409, 2004.
  • Robert H. Sloan and Balázs Szörényi:
    Revising projective DNF in the presence of noise.
    In Proc. Kalmár Workshop on Logic and Computer Science, October 2003, pages 143 -- 152.
  • Robert H. Sloan, Balázs Szörényi, and György Turán:
    Projective DNF formulae and their revision.
    Learning Theory and Kernel Machines (16th COLT/Kernel 2003), Springer LNAI 2777, 625 -- 639, 2003.
  • Judy Goldsmith, Robert H. Sloan, Balázs Szörényi, and György Turán:
    Improved algorithms for theory revision with queries.
    13th Annual Conference on Computational Learning Theory (13th COLT), 236 -- 247, 2000.


Bookchapters
  • Robert H. Sloan, Balázs Szörényi, and György Turán: Learning Boolean functions with queries. Boolean Models and Methods in Mathematics, Computer Science, and Engineering (ed. Peter L. Hammer and Yves Crama.), Cambridge University Press, 2010.