Valentin Neuhaus

Bachelor of Science in Physics, March 2021; Title: Analysis of Neural Network using Information Theory

Master of Science in Physics, September 2024; Title: Self-Organisation and Learning in Multilayer Infomorphic Networks

"The mechanistics of local-learning rules for curiosity in neural network models"

Within the RTG, my research explores modeling aspects of curiosity in neural networks through information-theoretic approaches and investigates whether local learning is sufficient to generate behavior driven by curiosity.

My research spans multiple interdisciplinary fields. I investigate neural networks through the frameworks of statistical physics and information theory, aiming to both understand and optimize neural computation. A particular focus is placed on self-organization and the emergent behaviors that arise within these systems.

• Schneider A.C., Neuhaus V., Ehrlich D.A., Makkeh A., Ecker A.S., Priesemann V., Wibral M. (2025). What Should a Neuron Aim For? Designing Local Objective Functions Based on Information Theory. International Conference on Learning Representations (ICLR)