Sandra Gilhuber

Supervision

I am constantly looking for talented and enthusiastic students for Master theses. You should have a basic understanding of Deep Learning and be familiar with writing clean and performant code in Python using NumPy and PyTorch.

Research Interests

• (Deep) Active Learning, Semi-Supervised Learning
• ML with Limited Labeled Data
• Domain Adaptation

Publications

• Sandra Gilhuber*, Rasmus Hvingelby*, Mang Ling Ada Fok, Thomas Seidl. "How To Overcome Confirmation Bias in Semi-Supervised Image Classification By Active Learning" Accepted at European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2023 (ECML PKDD 23)
• Sandra Gilhuber*, Julian Busch*, Daniel Rotthues, Christian Frey, Thomas Seidl. "DiffusAL: Coupling Active Learning with Graph Diffusion for Label-Efficient Node Classification" Accepted at European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2023 (ECML PKDD 23)
• Sandra Gilhuber, Philipp Jahn Yunpu Ma, Thomas Seidl.
  "Verips: Verified Pseudo-label Selection for Deep Active Learning"
  22nd IEEE International Conference on Data Mining (ICDM 2022)
• Sandra Gilhuber, Max Berrendorf, Yunpu Ma, Thomas Seidl.
  "Accelerating Diversity Sampling for Deep Active Learning By Low-Dimensional Representations"
  6th International Workshop on Interactive Adaptive Learning (IAL2022) Co-Located With ECML PKDD 2022
• Sandra Obermeier, Anna Beer, Florian Wahl, Thomas Seidl.
  "Cluster Flow — an Advanced Concept for Ensemble-Enabling, Interactive Clustering"
  19. Fachtagung für "Datenbanksysteme für Business, Technologie und Web" (BTW-21).
• Michael Fromm*, Max Berrendorf*, Sandra Obermeier, Thomas Seidl, and Evgeniy Faerman.
  "Diversity Aware Relevance Learning for Argument Search"
  43rd European Conference on Information Retrieval (ECIR-21)
• Sandra Obermeier, Max Berrendorf, Peer Kröger.
  "Memory-Efficient RkNN Retrieval by Nonlinear k-Distance Approximation"
  The 2020 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT'20)