DataMin seminars
2017 2016   2015   2014   2013   2012   2011   2010   2009   2008   2007
2012
22.12.2012
Friday
Metric Spaces and Positive Definite Functions
Presenter: Terkál Róbert
Links:
  • I.J. Schoenberg: Metric Spaces and Positive Definite Functions, Transactions of the American Mathematical Society, Vol. 44, No. 3 (Nov., 1938), pp. 522-536 PDF.

15:00, Math 207

07.12.2012
Friday
Efficient Learning and Feature Selection in High-Dimensional Regression
Presenter: Lehel Csató
Links:
  • Ting J-A, D'Souza A, Vijayakumar S, Schaal S (2010): Efficient Learning and Feature Selection in High-Dimensional Regression, Neural Computation, 22/4, 831-886. PDF.

15:00, Math 207

30.11.2012
Friday
Self-Taught Hashing for Fast Similarity Search
Presenter: Zalán Bodó
Links:
  • Dell Zhang and Jun Wang and Deng Cai and Jinsong Lu: Self-Taught Hashing for Fast Similarity Search. SIGIR'10, 18-25. PDF.

15:00, Math 207

23.11.2012
Friday
Kernel Based Least Squares Policy Iteration for Reinforcement Learning
Presenter: Hunor Jakab
Links:
  • Xin Xu, Dewen Hu, Xicheng Lu: Kernel-Based Least Squares Policy Iteration for Reinforcement Learning, IEEE Transactions on Neural Networks, 2007 PDF.

15:00, Math 207

16.11.2012
Friday
Experience Replay for Real-Time Reinforcement Learning Control
Presenter: Botond Bócsi
Links:
  • Sander Adam, Lucian Busoniu and Robert Babuska: Experience Replay for Real-Time Reinforcement Learning Control, IEEE Transactions on Systems, Man, and Cybernetics,2012 PDF.

15:00, Math 207

09.11.2012
Friday
A Few Useful Things to Know about Machine Learning
Presenter: Terkál Róbert
Links:
  • Pedro Domingos: A Few Useful Things to Know about Machine Learning, 2012 PDF.

15:00, Math 207

02.11.2012
Friday
Relative Entropy Policy Search
Presenter: Hunor Jakab
Links:
  • Jan Peters and Katharina Muelling and Yasemin Altun: Relative Entropy Policy Search, AAAI, 2012. PDF.

15:00, Math 207

26.10.2012
Friday
Kullback-Leibler Learning II
Presenter: Lehel Csató
Links:
  • Bierkens J, Kappen H.J (2012): KL-Learning: Online Solution of Kullback-Leibler Control Problems, (ArXiv, 2012) PDF.

15:00, Arcalia

19.10.2012
Friday
Kullback-Leibler Learning I
Presenter: Lehel Csató
Links:
  • Bierkens J, Kappen H.J (2011): Online Solution of the Average Cost Kullback-Leibler Optimisation Problems, (NIPS, 2011) PDF.

15:00, Arcalia

12.10.2012
Friday
Constraint satisfaction problems with isolated solutions are hard
Presenter: Terkál Róbert
Links:
  • Lenka Zdeborová, Marc Mézard (2008): Constraint satisfaction problems with isolated solutions are hard PDF.

15:00, Math207

05.10.2012
Friday
Deep belief nets
Presenter: Botond Bócsi
Links:
  • G.E. Hinton, S. Osindero, Y-W Teh (2006): A fast learning algorithm for deep belief nets PDF.

14:00, Math207

26.09.2012
Wednesday
Large-scale RLSC Learning Without Agony
Presenter: Zalán Bodó
Links:
  • W. Li, K-H Lee, K-S Leung (ICML 2007): Large-scale RLSC Learning Without Agony PDF.

15:00, Math207

19.09.2012
Wednesday
Surprise-based Sparsification
Presenter: Hunor Jakab
Links:
  • W. Liu, I Park, J. Principe (2009): An information-theoretic approach of designing sparse kernel adaptive filters, IEEE Transactions on Neural Networks 20/12, pp. 1950-1961, PDF.

15:00, Math207

12.09.2012
Wednesday
Representation in Continuous MDP's
Presenter: Lehel Csató
Links:
  • S. Mahadevan, M. Maggioni, K. Ferguson, S. Osentoski (2006): Learning Representation and Control In Continuous MDP's, AAAI Conference, PDF.

15:00, Math207

05.09.2012
Wednesday
Online Sparse Gaussian Process Regression
Presenter: Lehel Csató
Links:
  • A. Ranganathan, M-H Yang, J. Ho (2011) Online Sparse Gaussian Process Regression and Its Applications, PDF.

15:00, Math207

SUMMER BREAK

27.06.2012
Wednesday
Propagation of Uncertainty in Bayesian Kernel Models
Presenter: Hunor Jakab
Links:
  • J.Q. Candela, A. Girard, J. Larsen, C.E. Rasmussen (2004, TR) Propagation of uncertainty in bayesian kernel models - application to multiple-step ahead forecasting PDF.

15:00, Math207

20.06.2012
Wednesday
Survey Propagation -- algorithmics -- IV
Presenter: Lehel Csató
Links:
  • L. Kroc, A. Sabharwal, B. Selman (2007) Survey Propagation Revisited, Uncertainty in Artificial Intelligence. PDF.

15:00, Math207

13.06.2012
Wednesday
Survey Propagation -- studies and theory -- III
Presenter: Lehel Csató
Links:
  • A. Braunstein, M. Mezard, R. Zecchina (ArXiv 2006) Survey Propagation: an algorithm for satisfiability. PDF.
  • Michel Talagrand (2001) Riguruous results for mean field models for spin glasses, Probability Theory and Related Fields, 117(3), pp. 303-360. PDF.

15:00, Math207

06.06.2012
Wednesday
A New Look at Survey Propagation and Its Generalizations -- II
Presenter: Lehel Csató
Links:
  • Reading from section 4.

15:00, Math207

30.05.2012
Wednesday
A New Look at Survey Propagation and Its Generalizations
Presenter: Lehel Csató
Links:
  • Elitza Maneva, Elchanan Mossel, and Martin Wainwright (2007) ACM. PDF.

15:00, Math207

23.05.2012
Wednesday
Boosting k-Nearest Neighbor Classifier by Means of Input Space Projection
Presenter: Bócsi Botond
Links:
  • Nicolás García-Pedrajas, Domingo Ortiz-Boyer: Boosting k-Nearest Neighbor Classifier by Means of Input Space Projection, Expert Systems with Applications, 2009 . PDF.

15:00, Math207

16.05.2012
Wednesday
Learning geometric combinations of Gaussian kernels with alternating Quasi-Newton algorithm
Presenter: Hunor Jakab
Links:
  • David Picard, Nicolas Thome, Matthieu Cord, Alain Rakotomamonjy: Learning geometric combinations of Gaussian kernels with alternating Quasi-Newton algorithm, ESANN 2012. PDF.

15:00, Math207

9.05.2012
Wednesday
Derivative observations in Gaussian Process Models of Dynamic Systems
Presenter: Terkál Róbert
Links:
  • E. Solak, R. Murray-Smith, W.e. Leithead, D.J. Leith, and C.E. Rasmussen: Derivative observations in Gaussian Process models of dynamic systems, NIPS 2003. PDF.

15:00, Math207

18.04.2012
Wednesday
Music Plus One and Machine Learning
Presenter: Hunor Jakab
Links:
  • Christopher Raphael: Music Plus One and Machine Learning, ICML 2010. PDF.

15:00, Math207

04.04.2012
Wednesday
Optimization hardness as transient chaos in an analog approach to constraint satisfaction
Presenter: Lehel Csató
Links:
  • Mária Ercsey-Ravasz and Zoltán Toroczkai(2011): Optimization hardness as transient chaos in an analog approach to constraint satisfaction PDF.

15:00, Math207

28.03.2012
Wednesday
Mercer's Theorem, Feature Maps, and Smoothing
Presenter: Róbert Terkál
Links:
  • H.Q. Ming, P. Niyogi, Y. Yao (2006): Mercer's Theorem, Feature Maps, and Smoothing, COLT PDF.
    Presentation: PRESENTATION.

15:00, Math207

21.03.2012
Wednesday
Most Likely Heteroscedastic Gaussian Process Regression
Presenter: Lehel Csató
Links:
  • Kristian Kersting, Christian Plagemann, Patrick Pfaff, Wolfram Burgard (2007): Most Likely Heteroscedastic Gaussian Process Regression, ICML, 2007. PDF.

16:00, Math207

07.03.2012
Wednesday
Input noise in GP-s
Presenter: Botond A. Bócsi
Links:
  • McHutchon A, Rasmussen C.E (2011): Gaussian process Training with Input Noise, NIPS, 2011. PDF.

16:00, Math207

29.02.2012
Wednesday
Random Features
Presenter: Zalán Bodó
Links:
  • Ali Rahimi, Ber Rechts: Random Features for Large-Scale Kernel Machines, NIPS, 2007. PDF.

16:00, Math207

22.02.2012
Wednesday
Active Set Selection methods
Presenter: Hunor Jakab
Links:
  • Ricardo Henao and Ole Winther: Predictive Active Set Selection Methods for Gaussian Processes, International Workshop on Machine Learning for Signal Processing, 2010 PDF.

16:00, Math207

15.02.2012
Wednesday
Sparse Spectrum Gaussian Process Regression
Presenter: Lehel Csató
Links:
  • Lazaro-Gredilla, Miguel and Quinonero-Candela, Joaquin and Rasmussen, Carl Edward and Figueiras-Vidal, Anibal R.: Sparse Spectrum Gaussian Process Regression, Journal of Machine Learning Research, 2010 PDF.

16:00, Math207

08.02.2012
Wednesday
The Mathematics of Learning
Presenter: Lehel Csato
Links:
  • Tomaso Poggio and Steve Smale: The Mathematics of Learning (2003), Notices of the AMS PDF.

16:00, Math207

01.02.2012
Wednesday
Optimal Reinforcement Learning for Gaussian Systems
Presenter: Botond Bocsi
Links:
  • Philipp Hennig, Advances in Neural Information Processing Systems 24, 2011 PDF.

16:00, Math207

25.01.2012
Wednesday
Fast and accurate k-means for large data-sets
Presenter: Róbert Terkál
Links:
  • Braverman V, Meyerson A, Ostrovsky R, Roytman A, Shindler M, Tagiku B, Streaming k-means for well-clusterable data (SODA 2011) PDF.

16:00, Math207

18.01.2012
Wednesday
Fast and accurate k-means for large data-sets
Presenter: Zalán Bodó
Links:
  • Michael Shindler, Alex Wong, Fast and accurate k-means for large data-sets (NIPS 2011) PDF.

16:00, Math207

11.01.2012
Wednesday
Variational Heteroscedastic Gaussian Process Regression
Presenter: Botond Bócsi
Links:
  • Miguel Lazaro-Gredilla , Michalis Titsias, Variational Heteroscedastic Gaussian Process Regression (ICML 2011) PDF.

16:00, Math207