Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020068364 - ITERATIVE SOLVERS HAVING ACCELERATED CONVERGENCE

Publication Number WO/2020/068364
Publication Date 02.04.2020
International Application No. PCT/US2019/049002
International Filing Date 30.08.2019
IPC
G06F 17/12 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
17Digital computing or data processing equipment or methods, specially adapted for specific functions
10Complex mathematical operations
11for solving equations
12Simultaneous equations
G06F 17/15 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
17Digital computing or data processing equipment or methods, specially adapted for specific functions
10Complex mathematical operations
15Correlation function computation
G06N 3/02 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
CPC
G06F 17/12
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
17Digital computing or data processing equipment or methods, specially adapted for specific functions
10Complex mathematical operations
11for solving equations ; , e.g. nonlinear equations, general mathematical optimization problems
12Simultaneous equations ; , e.g. systems of linear equations
G06F 17/15
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
17Digital computing or data processing equipment or methods, specially adapted for specific functions
10Complex mathematical operations
15Correlation function computation ; including computation of convolution operations
G06N 3/0454
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
04Architectures, e.g. interconnection topology
0454using a combination of multiple neural nets
G06N 3/084
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
084Back-propagation
G06N 5/003
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
5Computer systems using knowledge-based models
003Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
Applicants
  • SIEMENS AKTIENGESELLSCHAFT [DE]/[DE]
  • THE BOARD OF TRUSTEES OF LELAND STANFORD JUNIOR UNIVERSITY [US]/[US]
Inventors
  • ERMON, Stefano
  • HSIEH, Jun-Ting
  • MIRABELLA, Lucia
  • ZHAO, Shengjia
  • EISMANN, Stephan
Agents
  • BRAUN, Mark, E.
Priority Data
62/737,14527.09.2018US
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) ITERATIVE SOLVERS HAVING ACCELERATED CONVERGENCE
(FR) SOLVEURS ITÉRATIFS AYANT UNE CONVERGENCE ACCÉLÉRÉE
Abstract
(EN)
Systems, methods, and apparatuses are described herein that learn fast iterative solvers that can be tailored to specific domains. Such fast iterative solvers are built on top of existing iterative solvers so as to converge at the correct solution while having enhanced convergence speeds as compared to existing solvers. In addition, such solvers can be applied to different geometries and boundary conditions as compared to the geometries and boundary conditions that are used to train the solvers.
(FR)
L'invention concerne des systèmes, procédés et appareils qui apprennent des solveurs itératifs rapides qui peuvent être adaptés à des domaines spécifiques. De tels solveurs itératifs rapides sont construits sur des solveurs itératifs existants de façon à converger au niveau de la solution correcte tout en ayant des vitesses de convergence améliorées par comparaison avec des solveurs existants. De plus, de tels solveurs peuvent être appliqués à différentes géométries et conditions limites par comparaison avec les géométries et conditions limites qui sont utilisées pour entraîner les solveurs.
Latest bibliographic data on file with the International Bureau