Paper Forecast NeurIPS 2024

Several BIFOLD research groups actively contribute to the program of the 48th Annual Conference on Neural Information Processing Systems (NeurIPS 2024), held in Vancouver, Canada, from December 10-15, 2024. The Machine Learning group of BIFOLD Co-Director Klaus-Robert Müller alone is represented at the conference with 8 puplications. NeurIPS covers a wide range of topics with machine learning and neuroscience, including cognitive science, psychology, computer vision, statistical linguistics, and information theory. 

BIFOLD research:

Machine Learning Group

1. Do Histopathological Foundation Models Eliminate Batch Effects? A Comparative Study
   1. Authors: Jonah Kömen, Hannah Marienwald, Jonas Dippel, Julius Hense
   2. Preprint: https://arxiv.org/abs/2411.05489
   3. Presented at: NeurIPS workshop “AIM-FM: Advancements In Medical Foundation Models: Explainability, Robustness, Security, and Beyond”
   4. Presented by: Jonah Kömen, Hannah Marienwald, Julius Hense
2. Doob's Lagrangian: A Sample-Efficient Variational Approach to Transition Path Sampling
   1. Authors: Yuanqi Du, Michael Plainer, Rob Brekelmans, Chenru Duan, Frank Noé, Carla P. Gomes, Alán Aspuru-Guzik, Kirill Neklyudov
   2. Preprint: https://arxiv.org/abs/2410.07974
   3. Code: https://github.com/plainerman/variational-doob
   4. Presented by: Michael Plainer
3. Federated Learning over Connected Modes
   1. Authors: Dennis Grinwald, Philipp Wiesner, Shinichi Nakajima
   2. Preprint: https://arxiv.org/abs/2403.03333
   3. Poster: https://nips.cc/virtual/2024/poster/95719
4. Generative Fractional Diffusion Models
   1. Authors: Gabriel Nobis, Maximilian Springenberg, Marco Aversa, Michael Detzel, Rembert Daems, Roderick Murray-Smith, Shinichi Nakajima, Sebastian Lapuschkin, Stefano Ermon, Tolga Birdal, Manfred Opper, Christoph Knochenhauer, Luis Oala, Wojciech Samek
   2. Preprint: https://arxiv.org/abs/2310.17638
5. MambaLRP: Explaining Selective State Space Sequence Models
   1. Authors: Farnoush Rezaei Jafari, Grégoire Montavon, Klaus-Robert Müller, Oliver Eberle
   2. Preprint: https://arxiv.org/abs/2406.07592
   3. Poster: https://nips.cc/virtual/2024/poster/96794
6. When Does Perceptual Alignment Benefit Vision Representations?
   1. Authors: Shobhita Sundaram, Stephanie Fu, Lukas Muttenthaler, Netanel Y. Tamir, Lucy Chai, Simon Kornblith, Trevor Darrell, Phillip Isola
   2. Preprint: https://arxiv.org/abs/2410.10817
   3. Blog post: https://ssundaram21.github.io/repalignment/
   4. Presented by: Shobhita Sundaram and Stephanie Fu
   5. Main conference paper
7.  xCG: Explainable Cell Graphs for Survival Prediction in Non-Small Cell Lung Cancer
   1. Authors: Marvin Sextro, Gabriel Dernbach, Kai Standvoss, Simon Schallenberg, Frederick Klauschen, Klaus-Robert Müller, Maximilian Alber, Lukas Ruff
   2. Preprint: https://arxiv.org/abs/2411.07643v1
   3. Code: https://github.com/marvinsxtr/explainable-cell-graphs/
   4. Presented at: Workshop “Machine Learning for Health (ML4H) symposium”
   5. Presented by: Marvin Sextro
8. xMIL: Insightful Explanations for Multiple Instance Learning in Histopathology
   1. Authors: Julius Hense, Mina Jamshidi Idaji, Oliver Eberle, Thomas Schnake, Jonas Dippel, Laure Ciernik, Oliver Buchstab, Andreas Mock, Frederick Klauschen, Klaus-Robert Müller
   2. Preprint: https://arxiv.org/abs/2406.04280
   3. Code: https://github.com/tubml-pathology/xMIL
   4. Presented by: Julius Hense, Mina Jamshidi Idaji

Further contributions linked to BIFOLD:

1. CoSy: Evaluating Textual Explanations of Neurons
   1. Authors: Laura Kopf, Philine Lou Bommer, Anna Hedström, Sebastian Lapuschkin, Marina M.-C. Höhne, Kirill Bykov
   2. Preprint: https://arxiv.org/abs/2405.20331
   3. Code: https://github.com/lkopf/cosy
   4. Presented by: Laura Kopf
2. Quanda: An Interpretability Toolkit for Training Data Attribution Evaluation and Beyond
   1. Authors: Dilyara Bareeva, Galip Ümit Yolcu, Anna Hedström, Niklas Schmolenski, Thomas Wiegand, Wojciech Samek, Sebastian Lapuschkin
   2. Preprint: https://arxiv.org/abs/2410.07158
   3. Presented at: ATTRIB Workshop
   4. Presented by: Galip Ümit Yolcu  
3. Breaking the curse of dimensionality in structured density estimation
   1. Authors: Robert A. Vandermeulen, Wai Ming Tai, Bryon Aragam
   2. Preprint: https://arxiv.org/abs/2410.07685
   3. Disclaimer: Much of this work was conducted at the Berlin Institute for the Foundations of Learning and Data (BIFOLD), Technische Universität Berlin.
4. Explainable AI needs formal notions of explanation correctness
   1. Authors: Stefan Haufe, Rick Wilming, Benedict Clark, Rustam Zhumagambetov, Danny Panknin, Ahcène Boubekki
   2. Preprint: https://arxiv.org/abs/2409.14590
   3. Presented by: Stefan Haufe
5. The effect of whitening on explanation performance
   1. Authors: Benedict Clark, Stoyan Karastoyanov, Rick Wilming, Stefan Haufe
   2. Presented by: Benedict Clark