AutoAI Seminar, Summer Term 2022

Material

Overview

This block seminar course will be held in English, towards the end of the 2022 summer semester. Enrollment is restricted to 20 Master students, preferably with a background in AI (including methods from machine learning, optimisation, planning and scheduling, multi-agent systems and other areas of AI). Students will work in groups of two on a range of topics from AutoAI, which spans the automated design and analysis of AI methods, including automated algorithm selection and configuration, automated machine learning techniques (including neural architecture search), and automated performance prediction. Each group will be assigned recently published work from the research literature, which will serve as the starting point for an in-depth investigation of a specific topic; the results of this investigation will be presented in class and compiled into a report.

Prerequisites

Preferably students should have a background in AI (including methods from machine learning, optimisation, planning and scheduling, multi-agent systems and other areas of AI).

Registration

Registration to the seminar is handled via the SuPra system.

Seminar topics

The seminar was hold in a two-day workshop in August 2022 [Workshop schedule].

  1. Topic: Algorithm selection in continuous optimization (AS-1)
    Students: Nils Sören Eberhardt, Stefan Seiler
  2. Topic: Algorithm selection for the discrete combinatorial problems (AS-2)
    Students: Nils Christoph Baumann, Tim Tobias Bauerle
  3. Topic: Local-search based algorithm configuration (AC-1)
    Students: Yimin Zhang, Marcel Baumann
  4. Topic: Sampling-based algorithm configuration (AC-2)
    Students: Jan Philipp Kraus, Nikolas Moritz Gunz
  5. Topic: Model-based algorithm configuration (AC-3)
    Students: Sebastian Miller, Laura-Sophie Kirchner
  6. Topic: Parallel algorithm portfolios (PAP)
    Students: Aaron Berger, Alexander Mann
  7. Topic: Gradient-Based Neural Architecture Search (NAS-1)
    Students: Marc Flemming Thiemann, Malte Gerhard Schwerin
  8. Topic: Evolutionary Computation for Neural Architecture Search (NAS-2)
    Students: Nicolas Maximilian Faber, Laszlo Dirks
  9. Topic: Empirical performance models (EPM)
    Students: Chenhuan Gao, Miriam Kempter

Organisers

Photo of Holger H. Hoos Prof. Dr. Holger H. Hoos Chair Holder, Alexander von Humboldt Professor

E-mail: hh[at]aim[dot]rwth-aachen[dot]de
Phone: +49 241 80 21451

Photo of Jakob Bossek Dr. Jakob Bossek Assistant Professor (Akademischer Rat)

E-mail: bossek[at]aim[dot]rwth-aachen[dot]de
Phone: +49 241 80 21453