Mittwoch, 27. Juni 2018

[HIForum] [Kolloquium] EINLADUNG / INVITATION - Informatisches Kolloquium 02.07.2018, 17:15, B201, Informatikum/Stellingen

Dies ist eine Einladung zum Informatischen Kolloquium am Montag, den 02. Juli 2018, um 17:15 auf dem Universitätsgelände "Informatikum/Stellingen", im Konrad-Zuse-Hörsaal B-201. Der Vortrag mit dem Titel "Evolution of Convolutional Highway Networks" wird von Herrn Prof. Dr. Oliver Kramer von der Universität Oldenburg gehalten (Vortragssprache: Deutsch/Englisch).

 

This is an invitation to the next informatics colloquium on Monday, 2 July 2018, at 17:15, on university campus Stellingen, lecture hall B-201. The talk is entitled "Evolution of Convolutional Highway Networks" and will be given in English or German (according to preference) by Prof. Dr. Oliver Kramer.

 

Nähere Informationen zu allen in diesem Semester geplanten Vorträgen finden Sie unter:

Please use the following link for more information on all talks planned in this semester:

https://www.inf.uni-hamburg.de/home/kolloquium/sose18.html

 

Im Namen des Kolloquium-Komitees

On behalf of the colloquium committee

Stephanie Schulte Hemming

Universität Hamburg

 

 

ABSTRACT:

Convolutional highways are based on multiple stacked convolutional layers for feature preprocessing. Like many other convolutional networks convolutional highways are parameterized by numerous hyperparameters that have to be tuned carefully. We introduce an evolutionary algorithm (EA) for optimization of the structure and tuning of hyperparameters of convolutional highways and demonstrate the potential of this optimization setting on the well-known MNIST data set. The EA employs Rechenberg's mutation rate control and a niching mechanism to overcome local optima. An experimental study shows that the EA is capable of evolving convolutional highway networks from scratch with only few evaluations but achieving competitive accuracy.

 

BIO:

Oliver Kramer's research interests are evolutionary optimization and deep learning with applications to real-world domains. He received a PhD from the University of Paderborn, Germany, in 2008. After postdoc stays at the TU Dortmund (Germany), Stanford and Berkeley (USA), he became professor for Computational Intelligence at the Department of Computing Science at the University of Oldenburg in 2011. He is the author of six books on evolutionary computation and machine learning.

  

CONTACT:

Prof. Dr. Mathias Fischer

 

 

Dienstag, 19. Juni 2018

[HIForum] [Kolloquium] EINLADUNG / INVITATION - Informatisches Kolloquium 25.06.2018, 17:15, B201, Informatikum/Stellingen

Dies ist eine Einladung zum Informatischen Kolloquium am Montag, den 25. Juni 2018, um 17:15 auf dem Universitätsgelände "Informatikum/Stellingen", im Konrad-Zuse-Hörsaal B-201. Der Vortrag mit dem Titel "An Integrative, Event-Predictive Perspective on Cognition: Behavioral Evidence and Artificial Neural Network Modelswird von Herrn Prof. Dr. Martin V. Butz von der Universität Tübingen gehalten (Vortragssprache: Englisch).

 

This is an invitation to the next informatics colloquium on Monday, 25 June 2018, at 17:15, on university campus Stellingen, lecture hall B-201. The talk is entitled "An Integrative, Event-Predictive Perspective on Cognition: Behavioral Evidence and Artificial Neural Network Models" and will be given in English by Prof. Dr. Martin V. Butz.

 

Nähere Informationen zu allen in diesem Semester geplanten Vorträgen finden Sie unter:

Please use the following link for more information on all talks planned in this semester:

https://www.inf.uni-hamburg.de/home/kolloquium/sose18.html

 

Im Namen des Kolloquium-Komitees

On behalf of the colloquium committee

Stephanie Schulte Hemming

Universität Hamburg

 

 

ABSTRACT:

I propose an integrative theory of cognition, derived from the principle of anticipatory behavior. Acknowledging that in the end all neural activities and encodings should serve homeostasis-oriented, behavioral control purposes – including abilities of adaptation, directing attention, social interaction including communication, versatile planning, and reasoning – behavior is controlled by desired future states. However, in order to make future state imaginations tractable, compact encodings of events and event transitions are essential. When augmenting formalizations of active inference – essentially minimizing anticipated free energy – with event-oriented abstractions, useful hierarchical, event-predictive encodings can develop. I show behavioral evidence that such encodings indeed exist and dynamically unfold in our minds. Moreover, I show several computational neuro-cognitive models that learn hierarchical, event-predictive encodings from sensorimotor experiences for the purpose of optimizing flexible and highly adaptive, interactive goal-directed behavior. I end with evidence that suggests that the joint activation of event-predictive states and successions of such may indeed make us language-ready. 

 

BIO:

Prof. Martin Butz is a full professor at the Department of Computer Science and the Department of Psychology at the University of Tübingen. While his main background lies in computer science (Diploma and PhD in Computer Science), and in machine learning in particular, he has been collaborating with cognitive psychologists, computational neuroscientists, roboticists, and linguists for many years. His research focusses on neuro-computational, cognitive modeling and cognitive science more generally. Currently, he is investigating the development of conceptual, cognitive language structures from sensorimotor experiences, pursuing an event-predictive approach. Moreover, he is relating these structures to the hippocampus, including the built-up of cognitive maps as well as the formation of episodic memory. He has published more than sixty peer-reviewed journal articles and more than one hundred peer-reviewed conference papers, as well as three monographs and seven edited volumes. The most recent monography, called "How the Mind Comes Into Being: Introducing Cognitive Science from a Functional and Computational Perspective" was published by Oxford University Press in 2017. Also, since 2017, Butz has been part of the Humboldt Network.

  

CONTACT:

Prof. Dr. Chris Biemann

 

 

Dienstag, 5. Juni 2018

[HIForum] [Kolloquium] INVITATION / EINLADUNG - Informatisches Kolloquium 11.06.2018, 17:15, B-201 Informatikum/Stellingen - Antrittsvorlesung

Dies ist eine Einladung zur Antrittsvorlesung von Prof. Dr. Peter Kling am Montag, den 11. Juni 2018, um 17:15 auf dem Universitätsgelände "Informatikum/Stellingen", im Konrad-Zuse-Hörsaal B-201. Der Titel lautet "Theorie Effizienter Algorithmen - Abstraktion & Algorithmische Puzzles" (auf Deutsch oder Englisch, je nach Wunsch der Zuhörer).

 

This is an invitation to the inaugural lecture of Prof. Dr. Peter Kling on Monday, 11 June 2018, 17:15, Campus "Informatikum/Stellingen", Room B-201, Konrad-Zuse lecture hall. The talk is entitled "Theory of Efficient Algorithms - On Algorithmic Puzzles & Real-World Problems" (in German or English according to preference).

 

Mehr Informationen zu allen in diesem Semester geplanten Vorträgen finden Sie unter:

The following link will provide information on all talks planned in this semester:

https://www.inf.uni-hamburg.de/home/kolloquium/sose18.html

 

In Namen des Kolloquium-Komitees/On behalf oft he Colloquium Committee

Stephanie Schulte Hemming

Universität Hamburg

 

 

ABSTRACT:

Komplexitätstheorie und effiziente Algorithmen sind ein zentraler Bestandteil der heutigen Informatik. Zur Grundausbildung jedes Informatikers gehören Methoden um algorithmische Lösungen für Optimierungsprobleme zu entwickeln und diese im Hinblick auf ihre Effizienz und Qualität zu analysieren. Wir werden darin geschult von realen Problemen zu abstrahieren, die inhärente Problemkomplexität heraus zu kondensieren und beweisbar gute Lösungsverfahren zu entwerfen – oder zu beweisen, dass das vorliegende Problem nicht effizient lösbar ist. Der Arbeitsbereich „Theorie Effizienter Algorithmen" (TEA) untersucht die inhärente Problemkomplexität in verschiedenen Bereichen, wie z.B. im verteilten Rechnen (wo wir uns mit dem Informationsfluss in großen, komplexen Netzwerken beschäftigen), in der Ressourcenverteilung in Rechenzentren und im Cloud Computing (wo die zur Verfügung stehenden Ressourcen unter wirtschaftlichen Gesichtspunkten verteilt werden müssen) oder auch in autonomen mobilen Systemen (wo wir untersuchen wie komplexe, globale Aufgaben durch eine Vielzahl einfacher, lokal agierender Agenten durchgeführt werden können). In meinem Vortrag werde ich verschiedene dieser Bereiche beleuchten und unsere aktuellen Forschungsfragen und -anstrengungen erläutern.

 

***

 

Complexity theory and efficient algorithms are a central pillar of modern computer science. Part of the basic education of any computer scientist is to learn to design algorithmic solutions for optimization problems and to reason formally about the efficiency and quality of these algorithms. We learn to look at a real-world problem, to distill and formalize the inherent difficulty in solving the problem, and to either design provably good solutions for them or to prove the impossibility of finding such solutions. The research group "Theory of Efficient Algorithms" (TEA) studies the inherent problem complexity in areas like distributed computing (where we deal with the information flow in huge, complex networks), resource management in data centers and in the cloud (where available resources must be distributed in an economic manner, even if we have incomplete information about the current or future system and workload characteristics), and autonomous mobile systems (where we study how simple agents with a locally restricted perception can cooperate to achieve a global goal). My talk will survey recent research efforts and progress in these areas.

 

 

BIO:

Peter Kling ist seit Januar 2018 Juniorprofessor an der Universität Hamburg. In seiner Forschung untersucht er fundamentale Eigenschaften sowie die Grenzen moderner Algorithmik. Zu seinen Kernforschungsgebieten gehören unter anderem verteilte Systeme, Online-Algorithmen, Ressourcenverteilung und stochastische Prozesse. Sein Diplom in der Mathematik und seinen Master in Informatik erhielt Herr Kling 2010 von der Universität Paderborn. Dort erhielt er 2014 auch den Doktorgrad für seine Arbeit zu energieeffizienten Algorithmen in der Ressourcenverteilung. Danach forschte er als Postdoc an der University of Pittsburgh (2014), im Rahmen eines Stipendiums des Pacific Institute for the Mathematical Sciences an der Simon Fraser University im Großraum Vancouver (2015 – 2016) sowie an der Universität Hamburg (2017).

 

***

 

Peter Kling is an assistant professor at the Universität Hamburg since January 2018. His research aims at exploring the fundamental properties and limits of algorithmic solutions to optimization and decision problems. His research interests include distributed systems, online algorithms, resource management, and stochastic processes. Peter received his diploma in Mathematics and his master's degree in Computer Science in 2010 from the Paderborn University, where he also received his PhD for his work on energy efficient scheduling algorithms in 2014. After that, he did postdocs at the University of Pittsburgh (2014), as a Fellow of the Pacific Institute for the Mathematical Sciences at the Simon Fraser University in Greater Vancouver (2015 – 2016), and at the University of Hamburg (2017).

  

 

CONTACT:

Prof. Tilo Böhmann