Gated attention neural networks

US12033055B2 · kind B2 · utility

Assignee

• DeepMind Technologies Limited · GB

Inventors

• Emilio Parisotto · Pittsburgh, US
• Hasuk Song · London, GB
• Jack William Rae · London, GB
• Siddhant Jayakumar · London, GB
• Maxwell Elliot Jaderberg · London, GB
• Razvan Pascanu · Letchworth, GB
• Caglar Gulcehre · Outwood, GB

Key dates

• Filing date: Sep 7, 2020
• Grant date: Jul 9, 2024
• Expiry date: Mar 23, 2041

Classification

• Technology area (CPC G): Physics
• CPC primary: G06N3/006
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

A system including an attention neural network that is configured to receive an input sequence and to process the input sequence to generate an output is described. The attention neural network includes: an attention block configured to receive a query input, a key input, and a value input that are derived from an attention block input. The attention block includes an attention neural network layer configured to: receive an attention layer input derived from the query input, the key input, and the value input, and apply an attention mechanism to the query input, the key input, and the value input to generate an attention layer output for the attention neural network layer; and a gating neural network layer configured to apply a gating mechanism to the attention block input and the attention layer output of the attention neural network layer to generate a gated attention output.