Technical Program (preliminary)

Monday, August 18; TIETOTALO, Room TB111

8:45 – 9:30

Registration

9:30 – 10:30

Plenary talk: JORMA RISSANEN

On optimal estimation

Jorma Rissanen (Tampere University of Technology, FINLAND)

10:30 – 11:00

Coffee break

11:00 – 12:00

Invited session: Information Theoretic Methods

Chair:

11:00 – 11:30

Entropy and the ‘compound’ law of small numbers

Ioannis Kontoyiannis (Athens University of Economics & Business, GREECE), O.T. Johnson (University of Bristol, UK), and M. Madiman (Yale University, USA)

11:30 – 12:00

Recent advances in computing the NML for discrete Bayesian networks

Petri Myllymäki (University of Helsinki, FINLAND)

12:00 – 13:30

Lunch break

13:30 – 14:30

Plenary talk: IMRE CSISZÁR

Minimization of entropy functionals under moment constraints

Imre Csiszár (Hungarian Academy of Sciences, HUNGARY)

14:30 – 15:50

Invited session: ITM in Engineering 1

Chair:

14:30 – 14:50

Normalized Maximum Likelihood models with memory for genomics

Ioan Tăbuş (Tampere University of Technology, FINLAND) and Jaakko Astola (Tampere University of Technology, FINLAND)

14:50 – 15:10

MDL and wavelet denoising with soft thresholding

Janne Ojanen (Helsinki University of Technology, FINLAND) and Jukka Heikkonen (European Commission - Joint Research Centre, ITALY)

15:10 – 15:30

On sequentially normalized maximum likelihood models

Teemu Roos (University of Helsinki, FINLAND) and Jorma Rissanen (University of Helsinki, FINLAND)

15:30 – 15:50

On the use of the Fisher information in computing the stochastic complexity

Ciprian Doru Giurcăneanu (Tampere University of Technology, FINLAND)

15:50 – 16:30

Coffee break

16:30 – 17:30

Plenary talk: MARCELO WEINBERGER

Information theory in an industrial research lab

Marcelo Weinberger (Hewlett-Packard Laboratories, USA)

18:00 – 20:00

Dinner

Tuesday, August 19; TIETOTALO, Room TB111

9:00 – 10:00

Invited session: ITM in Engineering 2

Chair:

9:00 –  9:30

Information optimization in early vision

Vijay Balasubramanian (University of Pennsylvania, USA)

 9:30 – 10:00

Estimation theory and information geometry based on denoising

Aapo Hyvärinen (University of Helsinki, FINLAND)

10:00 – 10:30

Coffee break

10:30 – 12:30

Poster session

P1

Optimized multiple description scalar quantization based 3D mesh coding

Mehmet Oguz Bici (Middle East Technical University, TURKEY), Gozde Bozdagi Akar (Middle East Technical University, TURKEY), Andrey Norkin (Tampere University of Technology, FINLAND), and Atanas Gotchev (Tampere University of Technology, FINLAND)

P2

Benchmarking of compression and speed performance for lossless audio compression algorithms

Florin Ghido (Tampere University of Technology, FINLAND) and Ioan Tăbuş (Tampere University of Technology, FINLAND)

P3

Testing goodness-of-fit via rate distortion

Peter Harremoës (Centrum voor Wiskunde en Informatica, THE NEDERLANDS)

P4

MDL and signal change detection in machine condition monitoring

Jenni Hulkkonen (Helsinki University of Technology, FINLAND) and Jukka Heikkonen (European Commission - Joint Research Centre, ITALY)

P5

NML-optimal histogram density estimation

Petri Kontkanen (University of Helsinki, FINLAND) and Petri Myllymäki (University of Helsinki, FINLAND)

P6

MDL knot selection for penalized splines

Antti Liski (Tampere University of Technology, FINLAND) and Erkki Liski (University of Tampere, FINLAND)

P7

Efficient computation of the multinomial NML

Tommi Mononen (University of Helsinki, FINLAND) and Petri Myllymäki (University of Helsinki, FINLAND)

P8

An NML-based method for learning Bayesian networks

Tomi Silander (University of Helsinki, FINLAND), Teemu Roos (University of Helsinki, FINLAND), and Petri Myllymäki (University of Helsinki, FINLAND)

P9

Calculating the NML distribution for tree-structured Bayesian networks

Hannes Wettig (University of Helsinki, FINLAND), Petri Kontkanen (University of Helsinki, FINLAND), and Petri Myllymäki (University of Helsinki, FINLAND)

P10

Asymptotic analysis of multiple description lattice vector quantization

Guoqiang Zhang (KTH Royal Institute of Technology, SWEDEN), Janusz Klejsa (KTH Royal Institute of Technology, SWEDEN), and W. Bastiaan Kleijn (KTH Royal Institute of Technology, SWEDEN)

12:30 – 13:30

Lunch break

13:30 – 14:30

Plenary talk: ANDREW BARRON

MDL procedures with ℓ₁ penalty and their statistical risk

Andrew Barron (Yale University, USA) and Xi Luo (Yale University, USA)

14:45 – 21:30

FREE DISCUSSIONS at the sauna Maisansalo by Näsijärvi lake

14:45 – 15:30

15:30 – 18:00

18:00 – 19:30

19:30 – 20:15

19:30 – 21:00

21:00 – 21:30

Bus from Hervanta to Maisansalo

Sauna (snacks + beer + non-alcoholic drinks)

Dinner (buffet)

(a) Bus to Hervanta and hotels

(b) Boat (Pyynikki 2) to Särkänniemi

(b) Bus from Särkänniemi to hotels and Hervanta

Wednesday, August 20; TIETOTALO, Room TB111

9:00 – 10:00

Plenary talk: SERGIO VERDU

Divergence

Sergio Verdu (Princeton University, USA)

10:00 – 10:30

Coffee break

10:30 – 12:00

Invited session: ITM in Engineering 3

Chair:

10:30 – 11:00

An overview of the use of distributed source coding in multimedia security

Nitin Khannan (Purdue University, USA) and Edward Delp (Purdue University, USA)

11:00 – 11:30

The use of ideas and methods of information theory to study communication and numerical  competence in ants

Boris Ryabko (Siberian State University of Telecommunications and Informatics, RUSSIA) and Zhanna Reznikova (Novosibirsk State University, RUSSIA)

11:30 – 12:00

Spatially adaptive support as a leading model-selection tool for image filtering

Vladimir Katkovnik (Tampere University of Technology, FINLAND), Alessandro Foi (Tampere University of Technology, FINLAND), Kostadin Dabov (Tampere University of Technology, FINLAND), and Karen Egiazarian (Tampere University of Technology, FINLAND)

12:00 – 13:30

Lunch break

13:30 – 15:30

Invited and contributed session: ITM in Statistics

Chair:

13:30 – 13:50

Model selection in linear mixed models using MDL criterion with an application to spline smoothing

Erkki Liski (University of Tampere, FINLAND) and Antti Liski (Tampere University of Technology, FINLAND)

13:50 – 14:10

Bayesian model selection

Arto Luoma (University of Tampere, FINLAND)

14:10 – 14:30

Modelling Markovian migration in finance and medicine

Rafael Weißbach (Dortmund University of Technology, GERMANY)

14:30 – 14:50

Partial correlation estimates based on signs

Daniel Vogel (Dortmund University of Technology), Claudia Köllmann (Dortmund University of Technology), and Roland Fried (Dortmund University of Technology)

14:50 – 15:20

Coffee break