Offentlige arrangementer

Extra Friday Lecture: Carolyn L. Talcott

Fault Tolerance From a Cellular Biology Perspective

Abstract:
We will give a high level overview of a wide range of `fault tolerance' mechanisms at the cellular level of biological systems.  We will then consider `fault tolerance' mechanismes in the immune system in more 
detail.  We will consider issues such as;
 
- perceived `goals' of the immune system

- what does fault tolerance mean in this setting

- underlying biological principles

- strategies used to provide defense against intrusion

- delicate balances that must be maintained

and relate these to security and fault tolerance as understood in  computer systems.

Friday Lecture: Ole Østerby

Så'n gik det til da farfar var ung

For godt 50 år siden kom jeg til Aarhus Universitet som studerende, og jeg har haft nær tilknytning til stedet lige siden.

Jeg vil fortælle lidt om studenterlivet, universitetet og byen dengang, om en lille afstikker til USA, og om den spæde begyndelse til datalogi i Aarhus og Datalogisk Institut.

MADALGO Seminar: Morteza Monemizadeh, Uni Dortmund

Title: Coresets and Sketches for High Dimensional Subspace Approximation Problems

Abstract:

We consider the problem of approximating a set $P$ of $n$ points in $\\Re^d$ by a $j$-dimensional subspace under the $\\ell_p$ measure, in which we wish to minimize the sum of $\\ell_p$ distances from each point of $P$ to this subspace. More generally, the $F_q(\\ell_{\\mathfrak{p}})$-subspace approximation problem asks for a $j$-subspace that minimizes the sum of $q$th powers of $\\ell_{\\mathfrak{p}}$-distances to this subspace, up to a multiplicative factor of $(1+\\epsilon)$.

We develop techniques for subspace approximation, regression, and matrix approximation that can be used to deal with massive data sets in high dimensional spaces. In particular, we develop coresets and sketches, i.e. small space representations that approximate the input point set $P$ with respect to the subspace approximation problem. Among the results we propose a dimensionality reduction method for various clustering measures, strong coreset, ptas and streaming algorithms in bounded and unbounded precision model for $F_1(\\ell_2)$-subspace approximation in high-dimensional spaces.

Luhmann og Medierne

Auditorim 4 i Søauditorierne

Ph.d.-forsvar: Vidar Falkenberg

Med henblik på erhvervelsen af ph.d.-graden i Informations- og Medievidenskab forsvarer Vidar Falkenberg fredag d. 11. december sin afhandling: Netavisernes historie - avismediets evolution i mediematricens revolution.