CCM and CCN Researchers Awarded Outstanding Paper Award at the International Conference on Learning Representations
The International Conference on Learning Representations (ICLR) honored Center for Computational Neuroscience (CCN) research fellow Florentin Guth, CCN guest researcher Zahra Kadkhodaie, Center for Computational Mathematics (CCM) distinguished research scientist Stéphane Mallat and CCN director Eero P. Simoncelli with one of its five Outstanding Paper Awards at its 2024 meeting in Vienna. The award was given in recognition of their study of image diffusion models titled “Generalization in diffusion models arises from geometry-adaptive harmonic representations.”
The team’s award-winning paper “covers a critical missing piece of our understanding of visual generative models and will likely inspire future important theory work in this area,” the awards committee wrote. The committee further noted that the authors “provide an explanation of this phenomenon in terms of architectural inductive biases by making a connection to ideas from harmonic analysis via ‘geometry-adaptive harmonic representations.’”
ICLR is a premier gathering of professionals dedicated to advancing the branch of artificial intelligence called representation learning, more commonly called deep learning. The conference is globally renowned for presentations of cutting-edge research on all aspects of deep learning used in artificial intelligence, statistics and data science, as well as important application areas such as machine vision, computational biology, speech recognition, text understanding, gaming and robotics.
