Ruhr-University Bochum
Faculty of Computer Science
Machine Learning / Algorithmics
Universitätsstr. 150

44801 Bochum

 

Room:  MC 5.124

Tel:      +49 (0)234 32-23207

E-Mail: asja.fischer@rub.de

Office hours: By Arrangement

Curriculum Vitae

Before becoming a professor in Bochum I was an assistant professor at RUB, akademische Rätin (assistant professor) at Bonn university, and a post-doctoral researcher at the Montreal Institute for Learning Algorithms (MILA). Between 2010 and 2015, I was employed both at the Institute for Neural Computation at the Ruhr-University Bochum and the Department of Computer Science at the University of Copenhagen working on my PhD, which I defended in Copenhagen in 2014. Before, I studied Biology. Bioinformatics, Mathematics, and Cognitive Science at the Ruhr-University Bochum, the Universidade de Lisboa, and the University of Osnabrück.

In a broad perspective, my main research ambition is to understand the fundamental computational principles of learning that characterize intelligence. More specifically, my research interests are focussed on the development, analysis, and application of deep learning models and methods. I am particularly interested in analysing and developing probabilistic models and inference methods, investigating biologically plausible deep learning, and understanding the stochastic processes involved in the training and optimisation of neural networks and probabilistic models.

Publications

2024

[1]
J. Ricker, S. Damm, T. Holz, und A. Fischer, „Towards the detection of diffusion model deepfakes“, in Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications – (Volume 4), Rom, März 2024, S. 446–457. doi: 10.5220/0012422000003660.
[2]
S. Däubener, A. Fischer, T. Glasmachers, und S. Mandt, „Leveraging stochasticity to increase the robustness of artificial neural networks“, Universitätsbibliothek, Ruhr-Universität Bochum, Bochum, 2024. doi: 10.13154/294-12582.
[3]
J. Ricker, D. Assenmacher, T. Holz, A. Fischer, und E. Quiring, „AI-generated faces in the real world: a large-scale case study of twitter profile images“, in Proceedings of the 27th international symposium on research in attacks, intrusions and defenses, Padua, Italien, Sep. 2024, S. 513–530. doi: 10.1145/3678890.3678922.
[4]
S. Mavali, J. Ricker, D. Pape, Y. Sharma, A. Fischer, und L. Schönherr, „Fake it until you break it: on the adversarial robustness of AI-generated image detectors“, 2. Oktober 2024. https://arxiv.org/abs/2410.01574
[5]
M. Laszkiewicz, J. Ricker, J. Lederer, und A. Fischer, „Single-model attribution of generative models through final-layer inversion“, in Proceedings of the 41st International Conference on Machine Learning (ICML 2024), 2024, Bd. 235, S. 26007–26042. [Online]. Verfügbar unter: https://raw.githubusercontent.com/mlresearch/v235/main/assets/laszkiewicz24a/laszkiewicz24a.pdf
[6]
J. Ricker, D. Lukovnikov, und A. Fischer, „AEROBLADE: training-free detection of latent diffusion images using autoencoder reconstruction error“, 31. Januar 2024.
[7]
J. Ricker, D. Lukovnikov, und A. Fischer, „AEROBLADE: training-free detection of latent diffusion images using autoencoder reconstruction error“, in Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, Sep. 2024, Bd. 2024. doi: 10.1109/cvpr52733.2024.00872.
[8]
J. Frank u. a., „A representative study on human detection of artificially generated media across countries“, in 2024 IEEE Symposium on Security and Privacy (SP), San Francisco, Sep. 2024, S. 55–73. doi: 10.1109/sp54263.2024.00159.
[1]
M.-R. Amini, S. Canu, A. Fischer, T. Guns, P. K. Novak, und G. Tsoumakas, Hrsg., Machine learning and knowledge discovery in databases, Part IV: European conference, ECML PKDD 2022, Grenoble, France, September 19–23, 2022, proceedings. Cham: Springer Cham, 2023. doi: 10.1007/978-3-031-26412-2.
[2]
K. Maag und A. Fischer, „Uncertainty-weighted loss functions for improved adversarial attacks on semantic segmentation“, 22. Mai 2023.
[3]
K. Maag und A. Fischer, „Uncertainty-weighted Loss Functions for Improved Adversarial Attacks on Semantic Segmentation“, 26. Oktober 2023.
[4]
M. Laszkiewicz, J. Ricker, J. Lederer, und A. Fischer, „Single-model attribution of generative models through final-layer inversion“, 26. Mai 2023.
[5]
J. Frank u. a., „A representative study on human detection of artificially generated media across countries“, 10. Dezember 2023.
[6]
M.-R. Amini, S. Canu, A. Fischer, T. Guns, P. Kralj Novak, und G. Tsoumakas, Hrsg., Machine Learning and Knowledge Discovery in Databases. Cham: Springer Nature Switzerland, 2023. doi: 10.1007/978-3-031-26422-1.
[1]
S. Däubener und A. Fischer, „How sampling impacts the robustness of stochastic neural networks“, Apr. 2022. doi: 10.48550/arxiv.2204.10839.
[2]
N. Zengeler, T. Glasmachers, A. Fischer, und U. Handmann, „Transfer meta learning: Herausforderungen der Mustererkennung“, Universitätsbibliothek, Ruhr-Universität Bochum, Bochum, 2022. doi: 10.13154/294-9296.
[3]
F. Linsner, L. Adilova, S. Däubener, M. Kamp, und A. Fischer, „Approaches to uncertainty quantification in federated deep learning“, in Machine learning and principles and practice of knowledge discovery in databases, online, Feb. 2022, Bd. 1524, S. 128–145. doi: 10.1007/978-3-030-93736-2_12.
[4]
J. Ricker, S. Damm, T. Holz, und A. Fischer, „Towards the detection of diffusion model deepfakes“, 26. Oktober 2022. https://arxiv.org/abs/2210.14571
[1]
K. Brügge, A. Fischer, und C. Igel, „On the convergence of the Metropolis algorithm with fixed-order updates for multivariate binary probability distributions“, in International Conference on Artificial Intelligence and Statistics, Online, 2021, Bd. 130, S. 469–477. [Online]. Verfügbar unter: https://proceedings.mlr.press/v130/brugge21a.html
[2]
D. Lukovnikov, S. Däubener, und A. Fischer, „Detecting compositionally out-of-distribution examples in semantic parsing“, in Findings of the association for computational linguistics – findings of ACL: EMNLP 2021, Dez. 2021, S. 591–598. doi: 10.18653/v1/2021.findings-emnlp.54.
[3]
A. P. Raulf, S. Däubener, B. L. Hack, A. Mosig, und A. Fischer, „SmoothLRP: smoothing LRP by averaging over stochastic input variations“, in 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, online, Okt. 2021, S. 599–604. doi: 10.14428/esann/2021.es2021-139.
[1]
J. Frank, T. Eisenhofer, L. Schönherr, A. Fischer, D. Kolossa, und T. Holz, „Leveraging frequency analysis for deep fake image recognition“, 26. Juni 2020. https://arxiv.org/pdf/2003.08685.pdf
[2]
S. Däubener, L. Schönherr, A. Fischer, und D. Kolossa, „Detecting adversarial examples for speech recognition via uncertainty quantification“, in Cognitive intelligence for speech processing, Okt. 2020, S. 4661–4665. doi: 10.21437/interspeech.2020-2734.
[3]
J. Frank, T. Eisenhofer, L. Schönherr, A. Fischer, D. Kolossa, und T. Holz, „Leveraging frequency analysis for deep fake image recognition“, in 37th International Conference on Machine Learning (ICML 2020), Online, 2020, Bd. 119, S. 3205–3216.
[4]
S. Däubener, J. Frank, T. Holz, und A. Fischer, „Efficient calculation of adversarial examples for bayesian neural networks “, 2020. [Online]. Verfügbar unter: https://openreview.net/pdf?id=0KlHSyXxHDW
[1]
A. Kristiadi, S. Däubener, und A. Fischer, „Uncertainty quantification with compound density networks“, gehalten auf der Conference on Neural Information Processing Systems (NeurIPS), Vancouver, 13. Dezember 2019, Publiziert. [Online]. Verfügbar unter: https://casa.rub.de/en/research/publications/detail/uncertainty-quantification-with-compound-density-network-workshop-beitrag
[2]
S. Däubener, A. Kristiadi, und A. Fischer, „Infinite ensembles for uncertainty prediction“, gehalten auf der Women in Machine Learning, Vancouver, 9. Dezember 2019, Publiziert. doi: 10.48550/arxiv.1902.01080.
[3]
A. Kristiadi, S. Däubener, und A. Fischer, „uncertainty quantification with compound density networks“, Konferenz-Abstract, 2019. [Online]. Verfügbar unter: http://bayesiandeeplearning.org/2019/papers/57.pdf
[1]
B. Weghenkel, A. Fischer, und L. Wiskott, „Graph-based predictable feature analysis“, Machine learning, Bd. 106, Nr. 9–10, S. 1359–1380, Mai 2017, doi: 10.1007/s10994-017-5632-x.
[1]B. Weghenkel, A. Fischer, und L. Wiskott, „Graph-based predictable feature analysis“, Machine learning, Bd. 106, Nr. 9–10, S. 1359–1380, Mai 2017, doi: 10.1007/s10994-017-5632-x.
[1]
J. Melchior, A. Fischer, und L. Wiskott, „How to center deep Boltzmann machines“, Journal of machine learning research, Bd. 17, Art. Nr. 99, 2016.
[1]
A. Fischer und C. Igel, „A bound for the convergence rate of parallel tempering for sampling restricted Boltzmann machines“, Theoretical computer science, Bd. 598, S. 102–117, 2015, doi: 10.1016/j.tcs.2015.05.019.
[1]
A. Fischer und C. Igel, „Training restricted Boltzmann machines: An introduction“, Pattern recognition, Bd. 47, Nr. 1, S. 25–39, 2014, doi: 10.1016/j.patcog.2013.05.025.
[1]
K. Brügge, A. Fischer, und C. Igel, „The flip-the-state transition operator for restricted Boltzmann machines“, Machine learning, Bd. 93, Nr. 1, S. 53–69, 2013, doi: 10.1007/s10994-013-5390-3.
[2]
J. Melchior, A. Fischer, N. Wang, und L. Wiskott, „How to center binary Restricted Boltzmann Machines“, 2013. [Online]. Verfügbar unter: http://arxiv.org/pdf/1311.1354v3.pdf
[3]
O. Krause, A. Fischer, T. Glasmachers, und C. Igel, „Approximation properties of DBNs with binary hidden units and real-valued visible units“, in 30th International Conference on Machine Learning, Atlanta, 2013, Bd. 28, S. 419–426.
[1]
A. Fischer und C. Igel, „An introduction to restricted Boltzmann machines“, in Progress in pattern recognition, image analysis, computer vision, and applications, Buenos Aires, 2012, Bd. 7441, S. 14–36. doi: 10.1007/978-3-642-33275-3_2.
[1]
A. Fischer und C. Igel, „Bounding the bias of contrastive divergence learning“, Neural computation, Bd. 23, Nr. 3, S. 664–673, 2011, doi: 10.1162/neco_a_00085.
[1]
A. Fischer und C. Igel, „Empirical analysis of the divergence of Gibbs sampling based learning algorithms for Restricted Boltzmann Machines“, in Artificial neural networks – ICANN 2010, Thessaloniki, 2010, Bd. 6352–6354, S. 208–217. doi: 10.1007/978-3-642-15825-4_26.

Publications

20 Einträge « 1 von 2 »
Artikel Geplante Veröffentlichung

Multiple Group Action Dlogs with(out) Precomputation

Alexander May, Massimo Ostuzzi

In: Preprint, Geplante Veröffentlichung.

Links | Schlagwörter: Preprint

Workshop

How to Lose Some Weight - A Practical Template Syndrome Decoding Attack

Sebastian Bitzer, Jeroen Delvaux, Elena Kirshanova, Sebastian Maaßen, Alexander May, Antonia Wachter-Zeh

Coding and Cryptography (WCC 24), 2024.

Links | Schlagwörter: Crypto Others

Proceedings Article

Too Many Hints - When LLL Breaks LWE

Alexander May, Julian Nowakowski

In: Advances in Cryptology (ASIACRYPT 23), 2023.

Links | Schlagwörter: Crypto Flagship, Rank A*/A

Proceedings Article

How to Enumerate LWE Keys as Narrow as in Kyber/Dilithium

Timo Glaser, Alexander May

In: Cryptology and Network Security (CANS 23), S. 75–100, Springer, 2023.

Links | Schlagwörter: Crypto Others

Proceedings Article

Breaking Goppa-based McEliece with hints

Elena Kirshanova, Alexander May

In: Security and Cryptography for Networks (SCN 22), and Journal of Information and Computation, Volume 293, 2023.

Links | Schlagwörter: Crypto Others

Proceedings Article

Low Memory Attacks on Small Key CSIDH

Jesús-Javier Chi-Dominguez, Andre Esser, Sabrina Kunzweiler, Alexander May

In: Applied Cryptography and Network Security (ACNS 23), S. 276–304, Springer, 2023.

Links | Schlagwörter: Crypto Others

Proceedings Article

New NTRU Records with Improved Lattice Bases

Elena Kirshanova, Alexander May, Julian Nowakowski

In: Post-Quantum Cryptography (PQCrypto 23), S. 167–195, Springer, 2023.

Links | Schlagwörter: Crypto Others

Proceedings Article

Dlog is Practically as Hard (or Easy) as DH - Solving Dlogs via DH Oracles on EC Standards

Alexander May, Carl Richard Theodor Schneider

In: Transactions on Cryptographic Hardware and Embedded Systems (TCHES), S. 146–166, 2023.

Links | Schlagwörter: Crypto Area, Rank A*/A

Proceedings Article

Partial Key Exposure Attacks on BIKE, Rainbow and NTRU

Andre Esser, Alexander May, Javier A. Verbel, Weiqiang Wen

In: Advances in Cryptology (CRYPTO 2022) , S. 346–375, Springer, 2022.

Links | Schlagwörter: Crypto Flagship, Rank A*/A

Proceedings Article

Approximate Divisor Multiples - Factoring with Only a Third of the Secret CRT-Exponents

Alexander May, Julian Nowakowski, Santanu Sarkar

In: Advances in Cryptology (EUROCRYPT 22) , S. 147–167, Springer, 2022.

Links | Schlagwörter: Crypto Flagship, Rank A*/A

20 Einträge « 1 von 2 »

Memberships

  • BITSI – Bochumer Verein zur Förderung der IT-Sicherheit und Informatik
  • CASA – DFG Excellence Cluster
  • QSI – EU Marie Curie Network
  • HGI – Horst Görtz Institute
  • IACR – Cryptology Research

Lectures (Moodle/Notes)

Former PhDs

  1. Önder Askin, 2024
  2. Floyd Zweydinger, 2023
  3. Lars Schlieper, 2022
  4. Alexander Helm, 2020
  5. Andre Esser, 2020
  6. Matthias Minihold, 2019 
  7. Leif Both, 2018
  8. Robert Kübler, 2018
  9. Elena Kirshanova, 2016
  10. Ilya Ozerov, 2016
  11. Gottfried Herold, 2014
  12. Alexander Meurer, 2014
  13. Mathias Herrmann, 2011
  14. Maike Ritzenhofen, 2010

Calvin & Hobbes

Prof. Dr. Asja Fischer

Machine Learning

Head of Chair

Address:
Ruhr-University Bochum
Faculty of Computer Science
Machine Learning / Algorithmics
Universitätsstr. 150
–D-44801 Bochum

Room: MC 5.124

Telephone: +49 (0)234 32 23207

Office Hours: By arrangement

E-mail: asja.fischer(at)rub.de

Curriculum Vitae

Before becoming a professor in Bochum I was an assistant professor at RUB, akademische Rätin (assistant professor) at Bonn university, and a post-doctoral researcher at the Montreal Institute for Learning Algorithms (MILA). Between 2010 and 2015, I was employed both at the Institute for Neural Computation at the Ruhr-University Bochum and the Department of Computer Science at the University of Copenhagen working on my PhD, which I defended in Copenhagen in 2014. Before, I studied Biology. Bioinformatics, Mathematics, and Cognitive Science at the Ruhr-University Bochum, the Universidade de Lisboa, and the University of Osnabrück.


Courses



Research

In a broad perspective, my main research ambition is to understand the fundamental computational principles of learning that characterize intelligence. More specifically, my research interests are focussed on the development, analysis, and application of deep learning models and methods. I am particularly interested in analysing and developing probabilistic models and inference methods, investigating biologically plausible deep learning, and understanding the stochastic processes involved in the training and optimisation of neural networks and probabilistic models.


Publications