notesum.ai

Published at November 12

Exact, Tractable Gauss-Newton Optimization in Deep Reversible Architectures Reveal Poor Generalization

cs.LG
cs.AI

Released Date: November 12, 2024

Authors: Davide Buffelli1, Jamie McGowan1, Wangkun Xu2, Alexandru Cioba1, Da-shan Shiu1, Guillaume Hennequin3, Alberto Bernacchia1

Aff.: 1MediaTek Research; 2Imperial College London; 3MediaTek Research & University of Cambridge

Arxiv: http://arxiv.org/abs/2411.07979v2