all AI news
Harnessing Orthogonality to Train Low-Rank Neural Networks
April 23, 2024, 4:44 a.m. | Daniel Coquelin, Katharina Fl\"ugel, Marie Weiel, Nicholas Kiefer, Charlotte Debus, Achim Streit, Markus G\"otz
cs.LG updates on arXiv.org arxiv.org
Abstract: This study explores the learning dynamics of neural networks by analyzing the singular value decomposition (SVD) of their weights throughout training. Our investigation reveals that an orthogonal basis within each multidimensional weight's SVD representation stabilizes during training. Building upon this, we introduce Orthogonality-Informed Adaptive Low-Rank (OIALR) training, a novel training method exploiting the intrinsic orthogonality of neural networks. OIALR seamlessly integrates into existing training workflows with minimal accuracy loss, as demonstrated by benchmarking on various …
abstract arxiv building cs.ai cs.lg dynamics investigation low multidimensional networks neural networks representation singular study svd train training type value
More from arxiv.org / cs.LG updates on arXiv.org
Jobs in AI, ML, Big Data
AI Research Scientist
@ Vara | Berlin, Germany and Remote
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
Data Science Analyst
@ Mayo Clinic | AZ, United States
Sr. Data Scientist (Network Engineering)
@ SpaceX | Redmond, WA