Inverse Problems are problems related to parameter and state estimation (inputs) based on (possibly) real-noisy observations (perturbed outputs). This kind of problems is widely occurred in many scientific fields. For instance, the application fields range from parameter estimation in Partial Differential Equations to state estimation in Data Assimilation. Usually, the estimation process is performed by making use of Bayesian inference wherein errors associated to priors and observations are assumed to follow some known probability distribution. Since the estimation is based on a stochastic process, commonly, the posterior estimate is chosen to be the Maximum A Posteriori (MAP) estimate, this is, the sample from the posterior distribution which maximizes the posterior probability. Depending on the number of components (parameters) to be estimated and the quality of the prior sample, the posterior estimate can provide meaningful or meaningless information about the true set of parameters and their corresponding uncertainty. This is clear since the MAP is nothing but a sample from the posterior distribution and depending on its unknown bias, for instance, it can be possible to improve the predicted parameters by taking another sample from the posterior distribution. This is a particular case which opens the door to stochastic algorithms in order to improve the quality of predicted values based on Bayesian inferences.



Papers are welcome on all aspects of Combinatorial Optimization applied to Inverse Problems, including, but not restricted to:



• Markov Chain Monte Carlo (MCMC) methods for computing posterior estimates.

• Combinatorial optimization in Bayesian inference.

• Local Search methods in Data Assimilation.

• Meta-heuristics in Covariance Matrix estimation.

• Stochastic Methods for Uncertainty Quantification.

• Stochastic Algorithms for Network Design.

• Decision Support Systems based on historical data.



Important Dates:



December 30, 2016: Expression of interest (title and abstract to guest editors)

February 28, 2017: Full manuscript and cover letter

April 30, 2017: Review comments and decision

May 30, 2017: Revised, final manuscript



Paper Submission System:



EasyChair [ https://easychair.org/conferences/?conf=ijaicomip2017 ]


Guest Editors:



Elias D. Nino-Ruiz, Ph.D.

Assistant Professor

Department of Computer Science

Universidad del Norte

Email: enino@uninorte.edu.co

Website: https://sites.google.com/a/vt.edu/eliasnino/

Barranquilla, Colombia



Xinwei Deng, Ph.D.

Associate Professor

Department of Statistics

Virginia Tech

Email: xdeng@vt.edu

Website: www.stat.vt.edu/people/faculty/Deng-Xinwei.html

Blacksburg, VA 24060, USA



Ivan Saavedra Antolínez, Ph.D.

Director of Professional Services

Competitive Insight, LLC

Email: isaavedra@ci-advantage.com

Website: https://www.linkedin.com/in/ivan-saavedra-antolinez-58111423/en

Atlanta, GA 30080, USA



Yezid Donoso, Ph.D.

Associate Professor

Department of Computer Science

Universidad del los Andes

Email: ydonoso@uniandes.edu.co

Website: https://sistemasacademico.uniandes.edu.co/~ydonoso/

Bogota, Colombia



-----------------------------



International Journal of Artificial Intelligence” does not want any publication fee for the International Journal of Artificial Intelligence.

But all authors of papers published in special issues need to pay a fee (which may be approx. 40 EUR) to download their paper after the publication in the journal.



----------------------------