Patent · US Active

Parameter utilization for language pre-training

US12072955B2 · kind B2 · utility

0Cited by
3References
20Claims
0Family size

Assignee

Inventors

Key dates

Filing dateNov 22, 2021
Grant dateAug 27, 2024
Priority date
Expiry dateFeb 15, 2043

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06F40/20
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

Embodiments are directed to pre-training a transformer model using more parameters for sophisticated patterns (PSP++). The transformer model is divided into a held-out model and a main model. A forward pass and a backward pass are performed on the held-out model, where the forward pass determines self-attention hidden states of the held-out model and the backward pass determines loss of the held-out model. A forward pass on the main model is performed to determine a self-attention hidden states of the main model. The self-attention hidden states of the main model are concatenated with the self-attention hidden states of the held-out model. A backward pass is performed on the main model to determine a loss of the main model. The parameters of the held-out model are updated to reflect the loss of the held-out model and parameters of the main model are updated to reflect the loss of the main model.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.