Uni-Logo

André Biedenkapp

André Biedenkapp

 PhD Student

Postal address:
Institut für Informatik
Albert-Ludwigs-Universität Freiburg
Sekretariat Hutter/Maschinelles Lernen
Georges-Köhler-Allee 074
79110 Freiburg, Germany
Room:
Building 074, Room 00-012
Coordinates:
48.014278, 7.831361 (DD)
Email:
biedenka@cs.uni-freiburg.de
Phone:
+49 761 203-54074
Fax:
+49 761 203-74217
Github:
AndreBiedenkapp
Twitter:
AndreBiedenkapp

About

Since October 2017 I am a PhD student at the Machine Learning Group under the supervision of Frank Hutter and Marius Lindauer. Before that I completed my master and bachelor degrees in computer science at the University of Freiburg.

Research Interests

I am interested in all facets of artificial intelligence. My research focuses on new ways to control the behavior of algorithms online. More precisely my research areas include:

Recent Publications

  • Biedenkapp, André and Rajan, Raghu and Hutter, Frank and Lindauer, Marius (published)(pdf)(arXiv)(code)(supplementary)(poster)(bib)
    TempoRL: Learning When to Act
    In: Proceedings of the 38th International Conference on Machine Learning (ICML 2021)
  • Eimer, Theresa and Biedenkapp, André and Hutter, Frank and Lindauer, Marius (published)(arXiv)(code)(poster)(bib)
    Self-Paced Context Evaluations for Contextual Reinforcement Learning
    In: Proceedings of the 38th International Conference on Machine Learning (ICML 2021)
  • Speck, David and Biedenkapp, André and Hutter, Frank and Mattmüller, Robert and Lindauer, Marius (published)(pdf)(arXiv)(code)(bib)(project page)
    Learning Heuristic Selection with Dynamic Algorithm Configuration
    In: Proceedings of the 31st International Conference on Automated Planning and Scheduling (ICAPS'21)

Publication Links

Code

  • Leading:
    • DAC: Dynamic algorithm configuration via reinforcement learning and accompanying white-box benchmarks.
    • PyImp: A python package to determine parameter importance based on configuration data.
    • CAVE: CAVE aims to help algorithm and configurator developers to better understand their experimental setup in an automated fashion. An example of a CAVE result can be found here. The video presentation for the corresponding CAVE paper as shown at LION'18 can be viewed here.
  • Involved:
    • DACBench: benchmark library for Dynamic Algorithm Configuration with focus on reproducibility and comparability of DAC approaches.
    • SMAC v3: Python implementation of SMAC (Algorithm Configurator) for automatic tuning of parameter configurations on any kind of algorithms.

Teaching

Misc

Erdös number: 4 (Holger H. Hoos -> Therese C. Biedl -> Jeffrey O. Shallit -> Paul Erdös)

My 3 word address: paid.laptops.brothers