VeranstaltungsortS.2.42Veranstalter Fakultät für Technische WissenschaftenBeschreibungScalable retrieval of high-dimensional feature vectors is an important component of many applications in multimedia and other fields, but also a very challenging problem. In this talk, we discuss the challenges of high-dimensional indexing at scale, and then present two approximate indexing methods designed for large-scale retrieval. We present results from experiments with the two largest feature collections reported in the literature, 28.5 billion SIFT features on a single server and 42.9 billion SIFT features in a distributed setting, and demonstrate an application with interactive retrieval over the 99.2 million images of the YFCC100M collection. Vortragende(r)Björn Thór Jónsson, Associate ProfessorIT University of Copenhagen, DenmarkReykjavik University, IcelandKontaktChristian Timmerer (christian.timmerer@itec.aau.at)

VeranstaltungsortHörsaal 3Veranstalter Institut für Innovationsmanagement und UnternehmensgründungBeschreibungKontaktInstitut für Innovationsmanagement und Unternehmen (innovation@aau.at)

Veranstaltungsorthttps://join.skype.com/fW14TGoEdd4ConlineVeranstalter Institut für Vernetzte und Eingebettete SystemeBeschreibungDear Uni-People,despite the Corona situation we continue to do research. As a special event I would like to invite everybody who is interested in the topics of energy management and technology to join our online meeting on the topic of energy forecasting. We have external and internal presentations on the topic.The meeting will take place on Wednesday, April 22, 2020 from 12:00-13:30. You can join the meeting using this link: https://join.skype.com/fW14TGoEdd4CAgenda:1. Welcome2. Short term load forecasting (Aida Mehdipourpirbazari, University of Stavanger, Norway)3. Machine learning techniques for power forecasting (Prof. Marco Mussetta, Politecnico di Milano)4. Towards a decentralized smart energy sharing: The ARTICONF approach (Nishant Saurabh)5. Investigating the impact of data quality on the energy yield forecast using data mining techniques (Ekanki Sharma)Vortragende(r)Aida MehdipourpirbazariProf. Marco MussettaNishant SaurabhEkanki SharmaKontaktWilfried Elmenreich (wilfried.elmenreich@aau.at)

Veranstaltungsort https://classroom.aau.at/b/anw-ezd-k9gVeranstalter Institut für MathematikBeschreibungTitel:Computation of eigenvalues for operators in an all-at-once formulationAbstract:The all-at-once formulation for inverse problems has recently been considered. The advantages of this approach include the avoidance of a parameter-to-state map and numerical improve- ments compared to the reduced setting. We want to find out about the eigenvalues of such operators formulated in an all-at-once fashion (i.e. putting the model and the observation equation in a joint model resulting in a block operator matrix). By means of an inverse source problem and the backward heat equation we perform the eigenvalue analysis both analytically and numerically. The operators of interest are transformed such that they are symmetric with respect to an L2-related inner product. It can be shown that the resulting operators lead to approximate eigenvalues, which can be computed in an analytic way. Finally, the problems are discretized and the analysis is done numerically.Vortragende(r)Anna Theresa SchlintlKontaktsenka haznadar (senka.haznadar@aau.at)