In-Context Learning of a Linear Transformer Block: Benefits of the MLP Component and One-Step GD Initialization

In-Context Learning of a Linear Transformer Block: Benefits of the MLP Component and One-Step GD Initialization

Item #:: 079017-0581
UPC:

Download PDF

Details

DOI: https://doi.org/10.52202/079017-0581
Author(s): Peter Bartlett, Jingfeng Wu, Ruiqi Zhang
Pages: 18310-18361 (52 pages)
Format: PDF Paper Download
Conference: Advances in Neural Information Processing Systems 37
Date/Location: Held 10-15 December 2024, Vancouver, Canada.
Series: Advances in Neural Information Processing Systems 37
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)

Description

Members/Attendees

Tab 4

Details
Description
Members/Attendees
Tab 4

DOI: https://doi.org/10.52202/079017-0581
Author(s): Peter Bartlett, Jingfeng Wu, Ruiqi Zhang
Pages: 18310-18361 (52 pages)
Format: PDF Paper Download
Conference: Advances in Neural Information Processing Systems 37
Date/Location: Held 10-15 December 2024, Vancouver, Canada.
Series: Advances in Neural Information Processing Systems 37
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)