NEWS
Publication Highlight – Pruning Clever-Hans strategies
Hidden Clever-Hans effects can undermine the reliability of AI models. The paper “Preemptively pruning Clever-Hans strategies in deep neural networks” introduces a method that corrects biases in neural networks without prior knowledge of faulty features.
A Dialogue of Art and Science
BIFOLD is happy to announce the Berlin-based artist duo kennedy+swan as Artists in Residence 2024/2025. Until June 2025, kennedy+swan will collaborate with BIFOLD researchers as part of the “Art of Entanglement” Artist in Residence program, which aims to bridge artistic and scientific perspectives on artificial intelligence.
EVENTS
AI and warfare
The objective of the conference is to explore these domains of modern warfare in order to develop a more accurate picture of the various effects of AI in military contexts. Another goal is to broaden the perspective of the military deployment of AI beyond questions of weapon systems and their control, by particularly looking at adversarial uses of AI in hybrid forms of warfare in the information domain.
BIFOLD Colloquium 11/2024
Join Prof. Enrico Magli as he explores how Edge-AI for Compression and Inference in Space is revolutionizing satellite data processing, featuring a novel compression algorithm and cutting-edge AI techniques.
Our Mission
BIFOLD conducts groundbreaking foundational research in Big Data Management (DM) and Machine Learning (ML), as well as their intersection, to educate the talents of the future and to create high-impact knowledge.
OPPORTUNITIES
BIFOLD welcomes job applications from interested parties year-round.
PUBLICATIONS
Robert A. Vandermeulen, Wai Ming Tai, Bryon Aragam
Breaking the curse of dimensionality in structured density estimation
Dilyara Bareeva, Galip Ümit Yolcu, Anna Hedström, Niklas Schmolenski, Thomas Wiegand, Wojciech Samek, Sebastian Lapuschkin
Quanda: An Interpretability Toolkit for Training Data Attribution Evaluation and Beyond
Adil Kabylda, J. Thorben Frank, Sergio Suarez Dou, Almaz Khabibrakhmanov, Leonardo Medrano Sandonas, Oliver T. Unke, Stefan Chmiela, Klaus-Robert Müller, Alexandre Tkatchenko
Molecular Simulations with a Pretrained Neural Network and Universal Pairwise Force Fields
GRADUATE SCHOOL
Based on a highly competitive application process, the BIFOLD Graduate School offers an innovative fast-track PhD Program for students holding a bachelor’s degree, as well as a PhD Program for students with a master’s degree.