Mareike Picklum

picklum_mareike.jpg
Tel: –49 -421 218 64016
Fax: –49 -421 218 64047
Room: TAB 1.77
Mail: mareikep(at)cs[dot]uni-bremen[dot]de

About

I studied Computer Science at the University of Bremen (UoB) and the University of New South Wales (UNSW) and received my Master's degree in 2015. I wrote my Master's thesis under the supervision of Prof. Michael Beetz at the Institute for Artificial Intelligence at the UoB addressing the problem of recognizing objects based on natural-language descriptions using graphical probabilistic models. I started my PhD in March 2015 in which I continue working with probabilistic models.

My work focuses on natural-language understanding and instruction interpretation to extract knowledge and use it to generate plans that can be executed by autonomous robotic systems.

I am a contributor in the projects pracmln and PRAC. In particular, I developed the publicly available web services PRACWeb, inviting users to have a look at the PRAC system in operation and WebMLN allowing users to try out inference and learning algorithms of the pracmln system.

Since February 2017 I am part of the initiative 'Farbige Zustände' (CRC 1232) which aims at the development of a novel experimental method for the development of materials. Within this collaborative research centre, I work on a prototype and feasibility study for an intelligent cognitive software assistant that supports the work of material scientists in designing new materials.

Teaching

  • AI: Knowledge Acquisition and Representation (WS2017/18) (Tutorial)
  • Foundations of Artificial Intelligence (SS2017) (Tutorial)
  • AI: Knowledge Acquisition and Representation (WS2016/17) (Tutorial)
  • Foundations of Artificial Intelligence (SS2016) (Tutorial)
  • AI: Knowledge Acquisition and Representation (WS2015/16) (Tutorial)
  • Foundations of Artificial Intelligence (SS2015) (Tutorial)

Supervised Theses

  • Crowdsourcing Instruction Data for Statistical Relational Learning (Bachelor's Thesis, Kevin Scheck)

Publications

Conference Papers
[1] Mihai Pomarlan, Daniel Nyga, Mareike Picklum, Sebastian Koralewski, Michael Beetz, "Deeper Understanding of Vague Instructions through Simulated Execution (Extended Abstract)", In Proceedings of the 2017 International Conference on Autonomous Agents & Multiagent Systems, International Foundation for Autonomous Agents and Multiagent Systems, 2017. [bib]
[2] Daniel Nyga, Mareike Picklum, Sebastian Koralewski, Michael Beetz, "Instruction Completion through Instance-based Learning and Semantic Analogical Reasoning", In International Conference on Robotics and Automation (ICRA), Singapore, 2017. [bib]
[3] Daniel Nyga, Mareike Picklum, Michael Beetz, "What No Robot Has Seen Before -- Probabilistic Interpretation of Natural-language Object Descriptions", In International Conference on Robotics and Automation (ICRA), Singapore, 2017. [bib]
Powered by bibtexbrowser