2015 Workshop on
Statistical Physics of Disordered Systems
and Its Applications (SPDSA2015)
--- Statistical-Mechanical Informatics and Statistical Machine Learning Theory in Big Data Sciences ---
Schedule: 19 February, 2015
Venue: Yoshida Izumidono, Yoshida Main/West Campus, Kyoto University, Yoshida-Izumidono-cho, Sakyo-ku, Kyoto, Japan
Yoshida Izumidono is Building No.76 in Yoshida Main/West Campus
(Access to Yoshida Izumidono in Google Map, Access to Kyoto).
Organized session only: We do not accept submissions from non-invited speakers.
10:00-12:00 Session 1
10:00-10:40 Kazuyuki Tanaka (Graduate School of Information Sciences, Tohoku University, Japan)
Title: Bayesian segmentation modeling by Potts prior and loopy belief propagation
In this talk, we present a Bayesian image segmentation modeling based on Potts prior and loopy belief propagation.
The proposed Bayesian model involves several terms, including the pairwise interactions of Potts models, and the average vectors and covariant matrices of Gauss distributions in color image modeling.
These terms are often referred to as hyperparameters in statistical machine learning theory.
In order to determine these hyperparameters, we propose a new scheme for hyperparameter estimation based on conditional maximization of entropy in the Potts prior.
The algorithm is given based on loopy belief propagation.
In addition, we compare our conditional maximum entropy framework with the conventional maximum likelihood framework, and also clarify how the first order phase transitions in loopy belief propagations for Potts models influence our hyperparameter estimation procedures.
K. Tanaka, S. Kataoka, M. Yasuda, Y. Waizumi, C.-T. Hsu: Bayesian image segmentations by Potts prior and loopy belief propagation, Journal of the Physical Society of Japan, vol.83, no.12, article no.124002, 2014.
K. Tanaka, S. Kataoka, M. Yasuda, M. Ohzeki: Inverse renormalization group transformation in Bayesian image segmentations, arXiv:1501.00834
10:40-11:20 Masayuki Ohzeki (Graduate School of Informatiocs, Kyoto University, Japan)
Title: Langevin dynamics violating detailed balanced condition
11:20-12:00 Keisuke Fujii (Graduate School of Informatics, Kyoto University, Japan)
Title: Belief propagation in quantum information science: a brief review and possible applications
14:30-15:30 Session 2
14:30-15:10 Botond Cseke (School of Informatics, University of Edinburgh, UK)
Title: Properties of Bethe free energies and message passing in Gaussian models
We address the problem of computing approximate marginals in Gaussian probabilistic models by using mean field and fractional Bethe approximations.
We define the Gaus- sian fractional Bethe free energy in terms of the moment parameters of the approximate marginals, derive a lower and an upper bound on the fractional Bethe free energy and establish a necessary condition for the lower bound to be bounded from below.
It turns out that the condition is identical to the pairwise normalizability condition, which is known to be a sufficient condition for the convergence of the message passing algorithm.
We show that stable fixed points of the Gaussian message passing algorithm are local minima of the Gaussian Bethe free energy.
By a counterexample, we disprove the conjecture stating that the unboundedness of the free energy implies the divergence of the message passing algorithm.
B. Cseke and T. Heskes: Properties of Bethe free energies and message passing in Gaussian models, vol.41, pp.1-24, 2011.
15:10-15:30 Arise Kuriya (Graduate School of Informatiocs, Kyoto University Japan)
Title: Approximate message passing algorithm in small scale problems
15:40-16:20 Session 3
15:40-16:20 Nial Friel (School of Mathematical Sciences, University College Dublin, Ireland)
Title: Convergence of Markov chains with approximate transition kernels ---applications to Markov random fields
Monte Carlo algorithms often aim to draw from a distribution π by simulating a Markov chain with transition kernel P such that π is invariant under P.
However, there are many situations for which it is impractical or impossible to draw from the transition kernel P.
For instance, this is the case with massive datasets, where is it prohibitively expensive to calculate the likelihood and is also the case for intractable likelihood models arising from, for example, Gibbs random fields, such as those found in spatial statistics and network analysis.
A natural approach in these cases is to replace P by an approximation P^.
Using theory from the stability of Markov chains we explore a variety of situations where it is possible to quantify how 'close' the chain given by the transition kernel P^ is to the chain given by P.
We apply these results to several examples from spatial statistics and network analysis.
P. Alquier, N. Friel, R. Everitt and A. Boland: Noisy Monte Carlo: Convergence of Markov chains with approximate transition kernels. Statistics and Computing (to appear), arXiv:1403.5496
16:20-17:00 Akihisa Ichiki (Green Mobility Collaborative Research Center, Nagoya University, Japan)
Title: Violation of detailed balance condition in Markov chaine Monte Carlo: From master equation to Langevin dynamics
17:00-17:20 Yuji Sakai (Graduate School of Arts and Sciences, University of Tokyo, Japan)
Title: Irreversible Markov chain Monte Carlo method with skew detailed balance condition
Registration Fee: Free
Organizers expect many reserachers and students in the related reserach fields
to attend the present workshop.
Masayuki Ohzeki (Graduate School of Informatics, Kyoto University, Japan)
Muneki Yasuda (Graduate School of Science and Engineering, Yamagata University, Japan)
Kazuyuki Tanaka (Graduate School of Information Sciences, Tohoku University, Japan)
``Foundations of Innovative Algorithms for Big Data'' in Research Area ``Advanced Core Technologies for Big Data Integration'' of JST-CREST, Japan
Graduate School of Information Sciences, Tohoku University, Japan
Workshop on Statistical Physics of Disordered Systems and Its Applications (SPDSA2013) --- Prologue Series V of FSPIP2013 --- (March, 2013, Sendai, Japan)
Frontier of Statistical Physics and Information Processing --- Perspectives from Nonequilibrium Behaviors --- (FSPIP2013) (July, 2013, Kyoto, Japan)
ELC International Meeting on ''Inference, Computation, and Spin Glasses'' (ICSG2013), (July, 2013, Sapporo, Japan)
Workshop on Statistical Physics of Disordered Systems and Its Applications (SPDSA2014) --- Inverse Problems and Statistical Machine Learning Theory --- (March, 2014, Kyoto, Japan)
Contact to SPDSA2015 office