Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020109774 - VERIFICATION OF PERCEPTION SYSTEMS

Publication Number WO/2020/109774
Publication Date 04.06.2020
International Application No. PCT/GB2019/053335
International Filing Date 26.11.2019
IPC
G06N 3/04 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
04Architecture, e.g. interconnection topology
G06N 5/00 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
5Computer systems using knowledge-based models
G06N 3/08 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
CPC
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 3/0472
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
0472using probabilistic elements, e.g. p-rams, stochastic processors
G06N 3/0481
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
0481Non-linear activation functions, e.g. sigmoids, thresholds
G06N 3/08
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
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
  • IMPERIAL COLLEGE OF SCIENCE, TECHNOLOGY AND MEDICINE [GB]/[GB]
Inventors
  • LOMUSCIO, Alessio
  • PANAGIOTIS, Kouvaros
Agents
  • THORNILEY, Peter
Priority Data
1819211.226.11.2018GB
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) VERIFICATION OF PERCEPTION SYSTEMS
(FR) VÉRIFICATION DE SYSTÈMES DE PERCEPTION
Abstract
(EN)
There is provided a computer-implemented method for verifying the robustness of a neural network classifier with respect to one or more parameterised transformations applied to an input, the classifier comprising one or more convolutional layers, the method comprising: encoding each layer of the classifier as one or more algebraic classifier constraints; encoding each transformation as one or more algebraic transformation constraints; encoding a change in an output classifier label from the classifier as an algebraic output constraint; determining whether a solution exists which satisfies the classifier constraints, transformation constraints and output constraints, and determining the classifier as robust to the local transformations if no such solution exists. A perception system and a computer readable medium are also provided.
(FR)
L'invention concerne un procédé mis en œuvre par ordinateur pour vérifier la robustesse d'un classificateur de réseau neuronal par rapport à une ou plusieurs transformations paramétrées appliquées à une entrée, le classificateur comprenant une ou plusieurs couches de convolution, le procédé consistant : à coder chaque couche du classificateur sous la forme d'une ou de plusieurs contraintes de classificateur algébriques ; à coder chaque transformation sous la forme d'une ou de plusieurs contraintes de transformation algébriques ; à coder un changement dans une étiquette de classificateur de sortie à partir du classificateur sous la forme d'une contrainte de sortie algébrique ; à déterminer si une solution existe qui satisfait les contraintes de classificateur, les contraintes de transformation et les contraintes de sortie, et à déterminer le classificateur comme robuste aux transformations locales si aucune pareille solution n'existe. L'invention porte également sur un système de perception et sur un support lisible par ordinateur.
Latest bibliographic data on file with the International Bureau