SETIA

We are pleased to announce the upcoming version of “Seminario de Teoría de la Información y sus Aplicaciones” SETIA. SETIA 2016 will take place on Wednesday, December 21, 2016, at room “Auditorio Principal”, A-Building, Universidad Técnica Federico Santa María, in Valparaíso, Chile.

This year’s program is the following:

10:00 to 11:00
“Una Introducción Intuitiva a la Teoría de la Información” (en español)
Slides
Dr. Milan S. Derpich
Associate Professor
Department of Eletronic Engineering,
Universidad Técnica Federico Santa María, Chile

Summary: En esta charla tutorial se redescubrirá, a partir de la
noción humana y cotidiana de información, las expresiones y cantidades
básicas utilizadas en la Teoría de la Información. Mediante un razonamiento
simple e intuitivo, se derivarán y discutirán las expresiones para la
entropía, la entropía condicional y la información mutua, y se dará un vistazo
a la propiedad de la equipartición asintótica y a algunas de sus
consecuencias. Finalmente se enunciarán y discutirán algunos de los resultados
más importantes de esta teoría, tales como le teorema de capacidad de canales
sin ruido, el de capacidad con canales con ruido, el teorema de la separación
y otros resultados asociados al compromiso entre tasa de datos y distorsión.

Summary: Claude Shannon is one of the leaders of the digital revolution that we are currently experiencing. James Clerk Maxwell, Ralph Hartley, John von Neumann, Alan Turing, and many other visionaries gave us computers capable of processing information. But it was Claude Shannon who introduced the mathematical theory of information in a 1948 article on which all modern communications are based. How to measure information? Digitization, transmission and storage of information, what are its fundamental limits? The answers lie in the legacy of Claude Shannon, of crucial importance for communications, computer science, statistics, cryptology and our digital society. In this talk we discuss these fundamental concepts, more than ever fruitful today.

14:00  to 14:40
“Astro-information, a new perspective on astronomical data using digital detectors”

Dr. Rene A. Mendez B.
Associate Professor
Astrominy Department  & National Astronomical Observatory
Universidad de Chile, Chile

Summary: Digital detectors have been in use for astronomical research since the early 80´s, with the inception of Charged-Coupled Devices (CCDs) for civil applications. Nowadays CCDs are the the most widely used detector at optical wavelengths in all major professional observatories around the globe – as well as in space-based observatories – and the expectation is that this type of detector will accompany us for several more years. Since CCD detectors have a well-characterized behavior, and motivated by the ever increasing measurement precision achieved by them, we have started to explore ways to characterize the maximum precision limits to astrometry (location) and photometry (brightness) that could possibly be achieved by these type of detectors under realistic experimental (observing) conditions. In this talk I will present some results of these efforts from our group. In recent years, the astronomical community has become used to the terms “astro-statistics” and “astro-informatics” to refer to the application of methodologies from these other areas of research (following the path of “astro-chemistry” and “astro-biology”). We believe that a better description of our research line will be “astro-information” (see ids-uchile.cl), which provides a fresher inter-disciplinary look into some old time problems.

14:40 PM to 15:20 PM
“The Causal Rate-Distortion Function for Gaussian Stationary Sources Under MSE Distortion”
Dr. Milan S. Derpich
Associate Professor
Department of Eletronic Engineering,
Universidad Técnica Federico Santa María, Chile

Summary: This talk will present a closed-form expression for the causal rate-distortion function (RDF) of Gaussian stationary sources under an average quadratic distortion (MSE) criterion. The RDF for Gaussian stationary sources under MSE distortion is known in closed form since the work of Kolmogorov in 1956. However, for the case in which the additional constraint of causality is imposed on the relationship between the source and its reconstruction, no explicit expressions are available, except for Gaussian i.i.d. or AR-1 sources or when the distortion goes to zero.
The closed-form expressions recently found by Derpich, Guerrero and Ostergaard are valid for any Gaussian stationary source and distortion. The talk will present and discuss this closed-form expression and take the audience through the main  insights  which led to this result.

Coffee Break

15:40 PM to 16:20 PM
“Almost Lossless Variable-Length Source Coding on Countably Infinite Alphabets”
Dr. Jorge F. Silva
Associate Professor
Department of Electrical Engineering/
Information and Decision Group

Facultad de Ciencias Físicas y Matemáticas
Universidad de Chile

Summary:  Motivated from the fact that universal source coding on countably infinite alphabets is not feasible, in this talk a notion of almost lossless source coding will be introduced.  This idea –analog to the  weak variable-length source coding proposed by Han 2000- aims at relaxing the lossless block-wise assumption to allow a distortion that vanishes asymptotically as the block-length goes to infinity.  In this almost lossless coding setting, new source coding results will be presented that on one hand show that Shannon entropy characterizes the minimum achievable rate (known statistics), while on the other,  that almost lossless universal source coding becomes feasible for the family of finite entropy stationary and memoryless sources with countably infinite alphabets.

Summary: In the first part of this talk, we investigate the problem of distributed biclustering of memoryless sources. This scenario consists of a set of distributed stationary memoryless sources where the goal is to find rate-limited representations such that the mutual information between two selected subsets of descriptions (each of them generated by distinct encoder functions) is maximized. This formulation is fundamentally different from conventional information-theoretic problems since here redundancy among descriptions should actually be maximally preserved. Furthermore, necessary and sufficient conditions for the special case of  two arbitrarily correlated Rademacher random variables and Boolean encoders are derived. Interestingly, these results positively resolve a long-standing open conjecture.

In the second part of the talk, we study the problem of collaborative distributed hypothesis testing. Two statisticians are required to declare the correct probability measure of two jointly distributed memoryless processes  out of two possible probability measures. The marginal samples given are assumed to be available at different locations and the statisticians are allowed to exchange limited amount of data over multiple rounds of interactions. A new achievable error exponent  is derived. Optimal achievable error exponents for the special cases of testing against independence and zero-rate communication (data exchanges grow sub-exponentially with n) are characterized.

SETIA 2016 is supported by the Advanced Center for Electrical and Electronic Engineering AC3E