← Search

David Dahmen

2 accepted papers

2024

Critical feature learning in deep neural networks

ICML 2024poster

A key property of neural networks driving their success is their ability to learn features from data. Understanding feature learning from a theoretical viewpoint is an emerging field with many open questions. In this work we capture finite-width effects with a systematic theory of network kernels in…

Cited by 3Source ↗PDF ↗Scholar ↗
2020

Unfolding recurrence by Green’s functions for optimized reservoir computing

NeurIPS 2020poster

Cortical networks are strongly recurrent, and neurons have intrinsic temporal dynamics. This sets them apart from deep feed-forward networks. Despite the tremendous progress in the application of deep feed-forward networks and their theoretical understanding, it remains unclear how the interplay of…

Cited by 5Source ↗PDF ↗Scholar ↗