Jilles Vreeken
530 posts
Check out our work on „Causal Discovery on Event Sequences by Local Cause-Effect Attribution“ on top of Whistler Mountain today, or, alternatively, Wednesday at 4:30pm in the West Hall #5104. Paper: eda.rg.cispa.io/prj/cascade/ Poster+Video: neurips.cc/virtual/2024/p…
Full paper and code now available! 😊 Follow the link to find out how we discover causal networks from data that arrives in episodes over time. 🌐 eda.rg.cispa.io/continent
🚨Our paper ”Learning Exceptional Subgroups by End-to-End Maximizing KL-divergence” received a spotlight at #ICML2024!🎉 Thank you @schaschaxu, @drjilles and Janis for this incredible collaboration. #ML4Science #interpretability #neurosymbolic 🧵1/7 openreview.net/forum?id=AG45X…
Does sharpness positively or negatively correlate with adversarial robustness? The answer is both! We explore this question and discover a peculiar property of the loss surface: "The Uncanny Valley". For more details, check out our paper arxiv.org/abs/2405.16918 .
🚨CFP #Causal #Representation #Learning @NeurIPSConf 2023 WS🚨 Do you have ideas on how to learn #causal representations from raw data, e.g. images, #abstract #macro variables or leverage #causality in #ML? Then submit a max 6 pg paper to our workshop crl-workshop.github.io
Now hiring: Three more PhD students and one postdoc for research on combining neural and symbolic/interpretable models of language, vision, and action. Work with world-class advisors at @Saar_Uni, MPI Informatics, @mpi_sws, @CISPA, @DFKI. Details: neuroexplicit.org/jobs
this is wild — kNN using a gzip-based distance metric outperforms BERT and other neural methods for OOD sentence classification intuition: 2 texts similar if cat-ing one to the other barely increases gzip size no training, no tuning, no params — this is the entire algorithm:
