Invention Application
- Patent Title: ENERGY-BASED ASSOCIATIVE MEMORY NEURAL NETWORKS
-
Application No.: US17441463Application Date: 2020-05-19
-
Publication No.: US20220180147A1Publication Date: 2022-06-09
- Inventor: Sergey Bartunov , Jack William Rae , Timothy Paul Lillicrap , Simon Osindero
- Applicant: DeepMind Technologies Limited
- Applicant Address: GB London
- Assignee: DeepMind Technologies Limited
- Current Assignee: DeepMind Technologies Limited
- Current Assignee Address: GB London
- International Application: PCT/EP2020/063971 WO 20200519
- Main IPC: G06N3/04
- IPC: G06N3/04

Abstract:
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for implementing associative memory. In one aspect a system comprises an associative memory neural network to process an input to generate an output that defines an energy corresponding to the input. A reading subsystem retrieves stored information from the associative memory neural network. The reading subsystem performs operations including receiving a given, i.e. query, input and retrieving a data element from the associative memory neural network that is associated with the given input. The retrieving is performed by iteratively adjusting the given input using the associative memory neural network.
Public/Granted literature
- US12277487B2 Energy-based associative memory neural networks Public/Granted day:2025-04-15
Information query