all AI news
Quadratic models for understanding neural network dynamics. (arXiv:2205.11787v1 [cs.LG])
May 25, 2022, 1:10 a.m. | Libin Zhu, Chaoyue Liu, Adityanarayanan Radhakrishnan, Mikhail Belkin
cs.LG updates on arXiv.org arxiv.org
In this work, we propose using a quadratic model as a tool for understanding
properties of wide neural networks in both optimization and generalization. We
show analytically that certain deep learning phenomena such as the "catapult
phase" from [Lewkowycz et al. 2020], which cannot be captured by linear models,
are manifested in the quadratic model for shallow ReLU networks. Furthermore,
our empirical results indicate that the behaviour of quadratic models parallels
that of neural networks in generalization, especially in the …
More from arxiv.org / cs.LG updates on arXiv.org
A Single-Loop Algorithm for Decentralized Bilevel Optimization
1 day, 10 hours ago |
arxiv.org
CLEANing Cygnus A deep and fast with R2D2
1 day, 10 hours ago |
arxiv.org
Jobs in AI, ML, Big Data
Data Architect
@ University of Texas at Austin | Austin, TX
Data ETL Engineer
@ University of Texas at Austin | Austin, TX
Lead GNSS Data Scientist
@ Lurra Systems | Melbourne
Senior Machine Learning Engineer (MLOps)
@ Promaton | Remote, Europe
Staff Software Engineer, Generative AI, Google Cloud AI
@ Google | Mountain View, CA, USA; Sunnyvale, CA, USA
Expert Data Sciences
@ Gainwell Technologies | Any city, CO, US, 99999