Workshop at NeurIPS 2024: Unifying Representations in Neural Models

On December 14, 2024, at NeurIPS, one of the most prestigious conferences in artificial intelligence, Zorah Lähner, Principal Investigator on Hybrid ML at the Lamarr Institute and head of the Geometry in Machine Learning at the University of Bonn, will co-organize the workshop Unifying Representations in Neural Models. This workshop will explore how and why different neural models—biological and artificial—tend to learn similar representations under comparable stimuli.

With contributions from experts in neuroscience, cognitive science, and AI, the workshop will cover topics such as learning dynamics, identifiability in functional spaces, and the practical applications of model alignment, including merging and reuse in multi-modal scenarios.

Speakers include Phillip Isola (MIT), Erin Grant (UCL), and Nina Miolane (UC Santa Barbara), among others. This event provides a unique platform to foster cross-disciplinary collaboration.