publications

2025

  1. A Survey on Archetypal Analysis
    Aleix Alcacer*Irene Epifanio*Sebastian Mair*, and Morten Mørup*
    arXiv preprint arXiv:2504.12392, 2025
  2. Personalized Privacy Amplification via Importance Sampling
    Dominik FaySebastian Mair, and Jens Sjölund
    Transactions on Machine Learning Research, 2025

2024

  1. Explaining the Impact of Training on Vision Models via Activation Clustering
    Ahcène BoubekkiSamuel G Fadel, and Sebastian Mair
    arXiv preprint arXiv:2411.19700, 2024
  2. Leveraging Activations for Superpixel Explanations
    Ahcène BoubekkiSamuel G Fadel, and Sebastian Mair
    arXiv preprint arXiv:2406.04933, 2024
  3. LoG
    Ising on the Graph: Task-specific Graph Subsampling via the Ising Model
    Maria Bånkestad, Jennifer Andersson, Sebastian Mair, and Jens Sjölund
    In The Third Learning on Graphs Conference, 2024
  4. Efficient radiation treatment planning based on voxel importance
    Sebastian Mair, Anqi Fu, and Jens Sjölund
    Physics in Medicine & Biology, Aug 2024
  5. Towards General Geometries for Embedding Knowledge Graphs
    Samuel G Fadel, Tino Paulsen, and Sebastian Mair
    In ICML/ELLIS Workshop on Geometry-grounded Representation Learning and Generative Modeling, Aug 2024
  6. Archetypal Analysis++: Rethinking the Initialization Strategy
    Sebastian Mair, and Jens Sjölund
    Transactions on Machine Learning Research, Aug 2024
  7. On Feynman–Kac training of partial Bayesian neural networks
    Zheng ZhaoSebastian MairThomas Schön, and Jens Sjölund
    In International Conference on Artificial Intelligence and Statistics, Aug 2024
  8. IDA
    Self-Supervised Siamese Autoencoders
    Friederike Baier, Sebastian Mair, and Samuel G Fadel
    In International Symposium on Intelligent Data Analysis, Aug 2024

2022

  1. Berechnung effizienter Datenzusammenfassungen
    Sebastian Mair
    In Ausgezeichnete Informatikdissertationen 2021, Aug 2022
  2. Studying the Propagation of Information in VAE Decoders
    Yannick Rudolph, Samuel G FadelSebastian Mair, and Ulf Brefeld
    In Northern Lights Deep Learning Conference, Aug 2022

2021

  1. thesis
    Computing Efficient Data Summaries
    Sebastian Mair
    Aug 2021
  2. AStA
    Contextual movement models based on normalizing flows
    Samuel G Fadel*Sebastian Mair*, Ricardo Silva Torres, and Ulf Brefeld
    AStA Advances in Statistical Analysis, Aug 2021
  3. Principled Interpolation in Normalizing Flows
    Samuel G Fadel*Sebastian Mair*, Ricardo da S Torres, and Ulf Brefeld
    In Joint European Conference on Machine Learning and Knowledge Discovery in Databases, Aug 2021

2020

  1. chapter
    Analyzing Positional Data
    Ulf Brefeld, Jan Lasek, and Sebastian Mair
    In Science meets Sports : when Statistics are more than Numbers, Aug 2020
  2. Efficient Normalizing Flows to Polytopes
    Sebastian MairSamuel G Fadel, Ricardo da S Torres, and Ulf Brefeld
    In Northern Lights Deep Learning Workshop, Aug 2020
  3. An Appropriate Prior Distribution for Interpolating Latent Samples in Flow-based Generative Models
    Samuel G FadelSebastian Mair, Ricardo da S Torres, and Ulf Brefeld
    In Northern Lights Deep Learning Workshop, Aug 2020

2019

  1. Coresets for Archetypal Analysis
    Sebastian Mair, and Ulf Brefeld
    In Advances in Neural Information Processing Systems, Aug 2019
  2. HyperUCB: Hyperparameter Optimization using Contextual Bandits
    Maryam TavakolSebastian Mair, and Katharina Morik
    In ECML-PKDD Workshop on Automating Data Science, Aug 2019
  3. Probabilistic movement models and zones of control
    Ulf Brefeld, Jan Lasek, and Sebastian Mair
    Machine Learning, Aug 2019

2018

  1. Distributed robust Gaussian process regression
    Sebastian Mair, and Ulf Brefeld
    Knowledge and Information Systems, Aug 2018
  2. Frame-Based Optimal Design
    Sebastian Mair, Yannick Rudolph, Vanessa Closius, and Ulf Brefeld
    In Joint European Conference on Machine Learning and Knowledge Discovery in Databases, Aug 2018
  3. Exploiting the Frame for Active Learning in Multi-class Classification
    Sebastian Mair, and Ulf Brefeld
    In ICML Workshop on Geometry in Machine Learning, Aug 2018

2017

  1. KDML
    Frame-based Matrix Factorizations
    Sebastian MairAhcène Boubekki, and Ulf Brefeld
    In LWDA Workshop on Knowledge Discovery, Data Mining and Machine Learning (KDML), Aug 2017
  2. Frame-based data factorizations
    Sebastian MairAhcène Boubekki, and Ulf Brefeld
    In International Conference on Machine Learning, Aug 2017

2015

  1. IHMMSEC
    Universal threshold calculation for fingerprinting decoders using mixture models
    Marcel Schäfer, Sebastian Mair, Waldemar Berchtold, and Martin Steinebach
    In Proceedings of the 3rd ACM Workshop on Information Hiding and Multimedia Security, Aug 2015