newsitem: ICML 2025
Our Data and AI cluster is happy to present no less than 8 contributions to the ICML 2025 conference:
Main track
Deep Neural Celullar Potts Models
Koen Minartz (TU/e - DAI), Tim d'Hondt (TU/e - DAI), Leon Hillmann (TU/e - DAI), Jörn Starruß (TU Dresden), Lutz Brusch (TU Dresden), Vlado Menkovski (TU/e - DAI)
Preference Controllable Reinforcement Learning with Advanced Multi-Objective Optimization
Yucheng Yang (TU/e - DAI), Tianyi Zhou (Uni Maryland), Meng Fang (Uni Liverpool, TU/e - DAI), Mykola Pechenizkiy (TU/e - DAI)
Bongard in Wonderland: Visual Puzzles that Still Make AI Go Mad?
Antonia Wüst (TU Darmstadt), Tim Tobiasch (TU Darmstadt), Lukas Helff (TU Darmstadt), Inga Ibs (TU Darmstadt), Wolfgang Stammer (TU Darmstadt), Devendra Singh Dhami (TU/e-DAI), Constantin A. Rothkopf (TU Darmstadt) and Kristian Kersting (TU Darmstadt)
LIFT the Veil for the Truth: Principal Weights Emerge after Rank Reduction for Reasoning-Focused Supervised Fine-Tuning
Zihang Liu (UC Berkeley) Tianyu Pang (Dartmouth College) Oleg Balabanov (UC Berkeley), Chaoqun Yang (Tsinghua University), Tianjin Huang (TU/e - DAI, University of Exeter) Lu Yin (TU/e - DAI, University of Surrey), Yaoqing Yang (Dartmouth College), Shiwei Liu (TU/e - DAI, University of Oxford)
Mask-Enhanced Autoregressive Prediction: Pay Less Attention to Learn More
Xialie Zhuang (University of Chinese Academy of Sciences; SCITIX (SGP) TECH PTE. LTD.), Zhikai Jia (SCITIX (SGP) TECH PTE. LTD.), Jianjin Li (South China Normal University), Zhenyu Zhang (University of Texas at Austin), Li Shen (Sun YatSen University), Zheng Cao (SCITIX (SGP) TECH PTE. LTD) Shiwei Liu (TU/e - DAI, University of Oxford)
From Low Rank Gradient Subspace Stabilization to Low-Rank Weights: Observations, Theories, and Applications
Ajay Jaiswal (University of Texas at Austin), Yifan Wang (Purdue University), Lu Yin (TU/e - DAI, University of Surrey) Shiwei Liu (TU/e - DAI, University of Oxford) Runjin Chen (University of Texas at Austin), Jiawei Zhao (6California Institute of Technology), Ananth Grama (Purdue University), Yuandong Tian (Meta), Zhangyang Wang (University of Texas at Austin)
Tracks and Workshops
The Interplay Between Implicit Bias and Adversarial Robustness in Linear Convolutional Neural Networks High Dimensional Learning Dynamics Workshop
Aurélien Boland (TU/e - DAI), Hannah Pinson (TU/e – DAI, VUB)
MEAL: A Benchmark for Continual Multi-Agent Reinforcement Learning Workshop on Multi-Agent Systems in the Era of Foundation Models
Tristan Tomilin (TU/e - DAI), Luka van den Boogaard (TU/e), Samuel Garcin (Uni Edinburgh), Bram Grooten (TU/e - DAI), Meng Fang(Uni Liverpool, TU/e - DAI), Mykola Pechenizkiy (TU/e - DAI)