Computer Science Engineering

Tivadar P pai - November *, ****

a

O ce Address:

Department of Computer Science

Information

University of Rochester

Rochester, NY 14627

Phone (o ce): 585-***-****

E-mail: *****@**.*********.***

WWW: www.cs.rochester.edu/ papai/

Markov logic, probabilistic graphical models, machine learning

Research

Recent projects:

interest

Preprocessing hard constraints with given evidence in Markov logic networks using

constraint programming techniques to reduce the size of the ground network and to

speed up the inference (joint work with Prof. Parag Singla (IIT, Delhi) and Prof.

Henry Kautz)

Introducing a Markov logic based probabilistic model in temporal/sequential do-

mains that has certain advantages over the standard conditional random eld based

Markov logic network instantiation (joint work with Prof. Henry Kautz and Prof.

Daniel Stefankovic)

Combining a domain expert s subjective probabilities with data to train Markov logic

networks within the framework of Bayesian statistics (joint work with Dr. Shalini

Ghosh (SRI International) and Prof. Henry Kautz)

Introducing modal operators in Markov logic to provide a framework for reasoning

under the principle of maximum entropy for modal logics K45, KD45 and S5 (joint

work with Prof. Henry Kautz and Prof. Daniel Stefankovic)

Work

experience

2011 Summer, Research Intern at SRI International, Menlo Park, California

Developing a weight learning algorithm under the supervision of Shalini Ghosh

which allows training of Markov logic networks when there is insu cient or no

training data available at all, but there is a domain expert whose knowledge can

be used to train the network

Implementing the weight learning algorithm in the Probabilistic Consistency En-

gine (PCE) in C

2010 Summer, Intern at IMO.IM, Palo Alto, California

Developing a VoIP application for Android and integrating it into the existing

imo.im Android application

2008 Summer, Intern at SAP Labs, Palo Alto, California

Implementing an SSL based tunneling tool in Java

Deploying BIRT into NetWeaver Application Server (J2EE)

University of Rochester, Rochester, New York USA

Education

Ph.D. in Computer Science, Department of Computer Science (GPA: 4.0/4.0)

Advisor: Professor Henry Kautz

M.S. in Computer Science, Department of Computer Science (graduated in 2008,

GPA: 4.0/4.0)

Budapest University of Technology and Economics, Budapest, Hungary

M.S. in Technical Informatics ( Computer Engineering), Faculty of Electrical En-

gineering and Informatics (graduated in June 2007 with excellent result, GPA:

4.71/5.00)

Thesis Topic: Intelligent Path Following Control of Model Vehicles

Advisor: Professor Laszl T. K czy

o o

Specialization: Software Engineering (Department of Automation and Applied

Informatics)

Participated in a Biomedical Engineering M.S. program for a semester studying

anatomy and biochemistry

Programming

Technical

Skills

Pro cient: Java

Experienced: C# .NET, C, Python

Basic level: Java EE, C++, Prolog, SML, Matlab

Past Research

experience

Intelligent Path Following Control of Model Vehicles

(prior

2006 spring - 2007 spring

University of

Writing the high level parts of the software in C# .NET

Rochester)

Developing a semicircle detection algorithm for position and orientation detection

Designing and implementing a controller for path following

Energy Balancing by Combinatorial Optimization for Wireless Sensor Networks

2005 fall - 2007 spring

SPECT: 2D - 3D Resolution Recovery (reducing collimator blurring by EM algo-

rithm)

2006 spring - 2006 fall

International

33rd International Physics Olympiad in Bali, Indonesia, 2002: Bronze

Competitions

medal

(296 students from 69 countries)

ACM International Collegiate Programming Contest, 2006

National Qualifying Contest 3rd place,

Central European Programming Contest, Honorable Mention

24-Hour Programming Contest, Final Round: 12th place (2nd place among Hungar-

ian teams)

Scholarships

GE (General Electric) Foundation Scholar-Leaders Program (2004-2006)

(awarded to 15 students each year in Hungary)

Scholarship of the Hungarian Republic (2006)

(awarded to the top 0.8% of the students at Budapest University of Technology and

Economics)

Publications

T. Papai, H. Kautz, D. Stefankovic. Reasoning Under the Principle of Maximum

Entropy for Modal Logics K45, KD45, and S5. To appear in Theoretical Aspects of

Rationality and Knowledge (TARK) 2013.

T. Papai, H. Kautz, D. Stefankovic. Slice Normalized Dynamic Markov logic Net-

works. To appear in Advances in Neural Information Processing Systems (NIPS)

2012.

T. Papai, S. Ghosh, H. Kautz. Combining Subjective Probabilities and Data in

Training Markov Logic Networks. European Conference on Machine Learning and

Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD 2012),

Bristol, UK, Septemeber 2012.

T. Papai, P. Singla, H. Kautz. Constraint Propagation for E cient Inference in

Markov Logic Networks. J. Lee (Ed.): Proceedings of the 17th International Con-

ference on Principles and Practice of Constraint Programming (CP 2011), LNCS

6876, pp. 691-705. Springer, Heidelberg, September 2011.

A. Geresdi, T. Papai, L. Siroki, L. T. Koczy. Component-based Hardware-software

System for Fuzzy Control of Automated Vehicles. In Proceedings of SEFI and IGIP

Joint Annual Conference 2007, pp. 397-399. Miskolc, Hungary, 1-4 July 2007.

ISBN 978-863-661-772-1

J. Levendovszky, A. Olah, Cs. Orosz, T. Papai, T. L. Tran. Energy Balancing by

Combinatorial Optimization for Wireless Sensor Networks. In Proceedings of the

6th International Workshop on Rare Event Simulation (RESIM 2006), pp. 74-85,

Bamberg, Germany, October 2006.

A. Geresdi, T. Papai, L. Siroki. In Atlas Series of the Solar System (10): Experimen-

tal Space Probe Model (Hunveyor Husar). Hungarian Academy of Sciences, Cosmic

Materials Space Research Group. Budapest, 2006. (Available only in Hungarian.)

ISBN 963 00 6314 X

ISBN 963 86873 6 3

Contact this candidate