Yuri Kalnishkan


Position 
Lecturer in Computer Science 
Email 
yura (at) cs (dot) rhul (dot) ac (dot) uk 
Phone 
(+44) 1784 41 4256 
Fax 
(+44) 1784 43 9786 
Research Area 
computational learning, online prediction, aggregating algorithm, predictive
and Kolmogorov complexity. 
CV 
in pdf
format (December 2015) 

Address
Department of Computer Science
Royal Holloway, University of London
Egham
Surrey
TW20 0EX
United Kingdom

Teaching
 Presessional mathematics classes for MSc
students. See the CS5100 moodle page for handouts.
 CS4200/CS5200, Online Machine Learning, Term 2.
 CS3930/CS5930, Computational Finance, Term 2.

PhD Students

Recent Talks

Articles on OnlinePrediction.net
Survey articles I have written for the wiki
project http://onlineprediction.net:

Video

Publications

Book Chapter
 Y.Kalnishkan. Predictive Complexity for Games with Finite Outcome
Spaces. In Measures of Complexity:
Festschrift for Alexey Chervonenkis, pp. 117139, Springer,
2015. DOI,
Pure.
In Journals
 T.Scarfe, W.Koolen, and Y.Kalnishkan. Segmentation of electronic
dance music. International Journal of Engineering Intelligent
Systems for Electrical Engineering and Communications. 22, 3/4
(2014).
Pure.
 Y.Kalnishkan, V.Vyugin, and V.Vovk. Generalised Entropies and Asymptotic Complexities of Languages. Information and Computation, 237, 101141 (2014). DOI, Pure.
 F.Zhdanov and Y.Kalnishkan. An Identity for Kernel Ridge
Regression. Theoretical Computer Science, 473, 157178
(2013). DOI, Pure.
 F.Zhdanov and Y.Kalnishkan. Universal Algorithms for Probability
Forecasting. International Journal on Artificial Intelligence
Tools, 21(4)
(2012). DOI, Pure.
 A.Chernov, Y.Kalnishkan, F.Zhdanov, and V.Vovk. Supermartingales
in Prediction with Expert Advice. Theoretical Computer Science,
411(2930):
26472669 (2010).
DOI,
Pure. See also the arXiv.org version.
 Y.Kalnishkan and M.V.Vyugin. The weak
aggregating algorithm and weak mixability. Journal of Computer
and System Sciences, 74(8): 12281244 (2008).
DOI,
Pure.
 Y.Kalnishkan, V.Vovk, and M.V.Vyugin. How
many strings are easy to predict? Information and
Computation, 201: 5571
(2005). DOI.
 Y.Kalnishkan, V.Vovk, and M.V.Vyugin. Loss
functions, complexities, and the Legendre transformation.
Theoretical Computer Science, 313(2): 195207,
(2004). DOI.
 Y.Kalnishkan. General linear relations
among different types of predictive complexity. Theoretical
Computer Science, 271(12): 181200, (2002).
DOI.

In Refereed Conference Proceedings
 T.Scarfe, W.Koolen, and Y.Kalnishkan. A longrange selfsimilarity approach to segmenting DJ mixed music streams. In Artificial Intelligence Applications and Innovations: Proceedings of the 9th IFIP WG 12.5 International Conference, AIAI 2013, Paphos, Cyprus, Springer, 2013. p. 235244.
DOI, Pure.
 F.Zhdanov and Y.Kalnishkan. An
Identity for Kernel Ridge Regression. In Algorithmic Learning
Theory 21st International Conference, ALT 2010, Proceedings,
volume 6331 of Lecture Notes in Computer Science, pages
405419. Springer, 2010.
DOI.
 F.Zhdanov and Y.Kalnishkan. Linear Probability
Forecasting. In Artificial Intelligence Applications and
Innovations, AIAI 2010, Proceedings, volume 339 of IFIP
Advances in Information and Communication Technology, pages
411. Springer, 2010.
DOI.
 A.Chernov, Y.Kalnishkan, F.Zhdanov, and V.Vovk. Supermartingales
in Prediction with Expert Advice. In
Algorithmic Learning Theory, 19th International
Conference, ALT 2008, Proceedings, volume 5254 of Lecture
Notes in Computer Science, pages 199213. Springer, 2008.
DOI.
 S.Busuttil and Y.Kalnishkan. Online
Regression Competitive with Changing Predictors. In Algorithmic
Learning Theory, 18th International Conference, ALT 2007,
Proceedings, volume 4754 of
Lecture Notes in Computer Science, pages 181195. Springer,
2007. DOI.
 S.Busuttil and Y.Kalnishkan. Weighted Kernel
Regression for Predicting Changing Dependencies. In Machine
Learning: ECML 2007, 18th European Conference on Machine Learning,
volume 4701 of Lecture Notes in Computer Science, pages
535542. Springer,
2007. DOI.
 Y.Kalnishkan, V.Vovk and
M.V.Vyugin. Generalised Entropy and Asymptotic
Complexities of Languages. In Learning Theory, 20th Annual
Conference on Learning Theory, COLT 2007, volume 4539
of Lecture Notes in Computer Science, pages 293307, Springer
2007. DOI.
The statement of the main theorem in this version of the paper is
inaccurate. See the journal version from Information and Computation,
2014 for the correct theorem.
 Y.Kalnishkan and M.V.Vyugin. The Weak
Aggregating Algorithm and Weak Mixability. In
Learning Theory, Proceedings of the 18th Annual Conference (COLT
2005), volume 3559 of Lecture Notes in Artificial
Intelligence, Springer, 2005.
DOI.
 Y.Kalnishkan, V.Vovk, and M.V. Vyugin. A
Criterion for the Existence of Predictive Complexity for Binary
Games. In Algorithmic Learning Theory, 15th International
Conference, ALT 2004, Proceedings, volume 3244 of Lecture Notes
in Artificial Intelligence, pages 249263. Springer, 2004.
DOI.
 A.Gammerman, Y.Kalnishkan, and
V.Vovk. Online Prediction with Kernels and the
Complexity Approximation Principle. In Uncertainty in
Artificial Intelligence, Proceedings of the Twentieth Conference,
pages 170176. AUAI Press, 2004.
ACM
Digital Library.
 Y.Kalnishkan, V.Vovk and M.V.Vyugin. How Many Strings Are Easy to
Predict? In 16th Annual Conference on Learning Theory (COLT) and
7th Annual Workshop on Kernel Machines, Proceedings, volume 2777
of Lecture Notes in Artificial Intelligence, SpringerVerlag,
2003. DOI.
 Y.Kalnishkan and M.V.Vyugin. On the Absence
of Predictive Complexity for Some Games. In Algorithmic
Learning Theory 13th International Conference, ALT 2002,
Proceedings,, volume 2533 of Lecture Notes in Artificial
Intelligence, SpringerVerlag, 2002.
DOI.
 Y.Kalnishkan and M.V.Vyugin. Mixability
and the Existence of Weak Complexities. In Computational
Learning Theory, 15th Annual Conference on Computational Learning
Theory, COLT 2002, Proceedings, volume 2375 of Lecture Notes in
Artificial Intelligence, pages 105120. Springer, 2002.
DOI.
 Y.Kalnishkan, M.V.Vyugin and V.Vovk. Loss Functions, Complexities,
and the Legendre Transformation. In Algorithmic Learning Theory
12th International Conference, ALT 2001, Proceedings, volume 2225
of Lecture Notes in Artificial Intelligence, pages
181189. SpringerVerlag, 2001.
DOI.
 Y.Kalnishkan. Complexity Approximation Principle and
Rissanen's Approach to RealValued Parameters. In Machine
Learning: ECML 2000, 11th European Conference on Machine Learning,
Proceedings, volume 1810 of Lecture Notes in Artificial
Intelligence, pages 203210, SpringerVerlag, 2000.
DOI.
 Y.Kalnishkan. General Linear Relations among Different Types of
Predictive Complexity. In Algorithmic Learning Theory, 10th
International Conference, ALT'99, Proceedings pages 323334,
volume 1720 of Lecture Notes in Artificial Intelligence,
SpringerVerlag,
1999. DOI.
 Y.Kalnishkan. Linear Relations between SquareLoss and Kolmogorov
Complexity. In Proceedings of the Twelfth Annual Conference
on Computation Learning Theory, pages 226232. Association
for Computing Machinery, 1999.

Technical Reports
 F.Zhdanov and Y.Kalnishkan. An Identity for Kernel Ridge Regression.
arXiv:1112.1390
 S.Busuttil, Y.Kalnishkan and A.Gammerman. Two New
Kernel Least Squares Based Methods for Regression. Technical Report CLRCTR0601,
Computer Learning Research Centre, Royal Holloway, University
of London, March 2006. Download: pdf.
 Y.Kalnishkan, V.Vovk, and M.V.Vyugin. A Criterion for the Existence
of Predictive Complexity for Binary Games. Technical Report CLRCTR0404,
Computer Learning Research Centre, Royal Holloway, University
of London, March 2004, revised May 2004. Download: postscript.
 Y.Kalnishkan and M.V.Vyugin. The Weak Aggregating Algorithm
and Weak Mixability. Technical Report CLRCTR0301, Computer
Learning Research Centre, Royal Holloway, University of London,
November 2003. Download: postscript.
 Y.Kalnishkan and V.Vovk. The existence of predictive complexity
and the Legendre transformation. Technical report CLRCTR0004,
Computer Learning Research Centre, Royal Holloway College, May
2000. Presented at TAI
2000, Fourth French Days on Algorithmic Information Theory.
Download: postscript.

See Also

Dissertation
The viva for the doctoral dissertation 'The Aggregating Algorithm and Predictive Complexity' took place on the 1st of October, 2002. Advisers: Volodya Vovk and Alex Gammerman. Examiners: Peter Gacs and Paul Vitanyi. Download: zipped postscript
(413 KB) or pdf (543 KB). The dissertation is also available on the ECCC (see this page for the abstract, table of contents, and another copy of the full text).

Research Grants
Date 
Grant 
Body 
20132016 
Grant RPG2013047 'Online selftuning learning algorithms for
handling historical information' 
Leverhulme Trust (Awards 2013) 
20072010 
Coinvestigator on the grant EP/F002998
'Practical competitive prediction' (with
Profs. V.Vovk and A.Gammerman) 
Engineering and Physical Sciences Research Council 
20012003 
Researcher Coinvestigator on the grant GR/R46670
'Complexity Approximation Principle and Predictive Complexity:
Analysis and Applications' (held by Profs. A.Gammerman and V.Vovk) 
Engineering and Physical Sciences Research Council 
19982001 
PhD funded by the grant GR/M14937
'Predictive Complexity: recursiontheoretic variants' (held
by Prof. V.Vovk) 
Engineering and Physical Sciences Research Council 

Academic Awards
Date 
Award 
Body 
2010 
Best paper award (with F.Zhdanov) 
6th IFIP Conference on
Artificial Intelligence Applications and Innovations, AIAI 2011

19982001 
Overseas Research Students Awards Scheme
grant 
Committee of ViceChancellors and Principals of
the Universities of the United Kingdom 
2000 
BrainBuster competition in MATLAB programming, first prize 
ECM/MathWorks 
1999 
E Mark Gold Award 
The program committee of the 10th International
Conference on Algorithmic Learning Theory (Tokyo, Japan) 

