All Years Seminars
[INMA] 2013-12-17 (14.00) : Low-rank Similarity based Extraction of Roles in Networks
At Euler building
Speaker :
Browet, Arnaud
Abstract :
In recent works, a lot of attention has been focused on the extraction of meaningful structures characterizing networks at aggregated levels.
Many researchers have proposed algorithms and methods to unfold community structures, i.e. groups of densely connected nodes, but little attention have been raised to extract more general structures known as roles. In this talk, we will present the problem of role modeling (a.k.a. Block modeling) and propose a pairwise similarity matrix that compares the flow pattern of each nodes in the network and allows the extraction of such role structures. Computing the exact similarity matrix being computationally difficult, we will present a low rank iterative scheme and demonstrate his convergence.
[INMA] 2013-11-28 (14.00) : Structure in Algorithmic Design for Identification and State Estimation
At Euler building
Speaker :
Michel Verhaegen (Delft Center for Systems and Control)
Abstract :
Mathematical models can be interpreted as structured relationship to represent (or approximate) physical dynamical phenomena. In this lecture we revisit how structure can be integrated in (1) subspace identification and (2) the solution of large scale receding horizon state reconstruction problems. In the first problem a solution is presented that integrates the separate algorithmic steps that so-far have characterized many subspace identification methods. It will be demonstrated that the integrative step enables to incorporate in an "elegant" manner structural information about the underlying system to be identified. A convex relaxation of the integrative solution is given. For the second problem the receding horizon estimation problem is used as illustration of a framework to design distributed algorithmic solutions for large scale systems and control problems. The lecture is concluded with a brief summary of the ERC project iCON.
[INMA] 2013-11-25 (14.00) : Recent advances in the field of respirometry for the characterization of biological processes
At Euler building
Speaker :
Frédéric Thalasso (Centro de Investigación y de Estudios Avanzados del Instituto Politécnico Nacional (Cinvestav), Mexico City)
Abstract :
Respirometry is the measurement of the biological oxygen uptake rate under well-defined conditions, commonly used to characterize biological processes. The interpretation of the respirograms; i.e. the respiration rate observed after a transient state has been applied to the biological system, allows for the determination of kinetic and stoichiometric parameters, used in process modeling. Respirometry is conceptually simple, but still complex to implement, as a correct appraisal of experimental data requires a mathematical approach and the understanding of the biological response to transient states.During this seminar, I will present the ongoing researches done at Cinvestav and in particular, the application of respirometry in microreactors, which allows for the generation of large data sets with relatively low experimental effort.
[INMA] 2013-11-12 (14.00) : Elements of a Nonstochastic Information Theory
At Euler building
Speaker :
Girish Nair (University of Melbourne, Australia)
Abstract :
In communications, unknown quantities are usually modelled as random variables. In contrast, control theory often treats uncertainties as bounded unknowns having no statistical structure. The area of networked control combines both fields and raises the question of whether it is possible to construct meaningful analogues of stochastic concepts such as independence, Markovness and information, without assuming a probability space. This talk describes a framework for doing so, leading in particular to the construction of a maximin information functional for non-stochastic variables. It is shown that, in this framework, the largest maximin information rate through a memoryless, error-prone channel coincides exactly with its block-coding zero-error capacity without feedback. This leads to a tight condition for the achievability of exponential uniform convergence when estimating the state of an unperturbed linear system over such a channel. Time permitting, recent results relating the concept of directed maximin information to zero-error feedback capacity and feedback control will also be discussed.
[INMA] 2013-10-29 (14.00) : Fundamental limits for noise suppression in biological networks
At Euler building
Speaker :
Ioannis Lestas (University of Cambridge)
Abstract :
The derivation of hard performance bounds in the presence of noisy feedback has been extensively addressed by control theorists over the last century, with many results going back to the early work of Wiener, Bode and Kalman. In a biochemical reaction network, however, where the noise is associated with the spontaneous births and deaths of individual molecules, many of the more conventional methodologies come to a halt, and need to be appropriately refined and extended. The approach in this talk is to address such complications by quantifying the way these spontaneous fluctuations restrict the ability to transmit information reliably, i.e. they lead to communication channels with finite capacity in the sense of Shannon, which can be explicitly calculated. We show that such restrictions in information transmission fundamentally limit the ability to suppress fluctuations in molecular concentrations, thus leading to hard bounds for noise suppression that hold for arbitrary feedback policies. The limits derived challenge conventional beliefs about biochemical accuracy, and provide an explanation to some striking experimental observations where a huge amount of resources are used for noise suppression within the cell. They also show how an interface between control and information theory can be used to rigorously analyze poorly characterized biological systems.
Seminars
Subscribe (ICS)
INMA Contact Info
Mathematical Engineering (INMA)
L4.05.01
Avenue Georges Lemaître, 4
+32 10 47 80 36
secretaire-inma@uclouvain.be
Mon – Fri 9:00A.M. – 5:00P.M.
JOIN US