WIPO logo
Mobile | Deutsch | Español | Français | 日本語 | 한국어 | Português | Русский | 中文 | العربية |
PATENTSCOPE

Search International and National Patent Collections
World Intellectual Property Organization
Options
Query Language
Stem
Sort by:
List Length
Some content of this application is unavailable at the moment.
If this situation persist, please contact us atFeedback&Contact
1. (WO2018046415) METHOD FOR CALCULATING A NEURON LAYER OF A MULTI-LAYER PERCEPTRON MODEL WITH SIMPLIFIED ACTIVATION FUNCTION
Latest bibliographic data on file with the International Bureau    Submit observation

Pub. No.: WO/2018/046415 International Application No.: PCT/EP2017/072042
Publication Date: 15.03.2018 International Filing Date: 04.09.2017
IPC:
G06N 3/063 (2006.01) ,G06N 3/04 (2006.01) ,G06N 7/00 (2006.01) ,G05B 13/04 (2006.01) ,F02P 5/15 (2006.01) ,G06F 15/78 (2006.01)
G PHYSICS
06
COMPUTING; CALCULATING; COUNTING
N
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3
Computer systems based on biological models
02
using neural network models
06
Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
063
using electronic means
G PHYSICS
06
COMPUTING; CALCULATING; COUNTING
N
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3
Computer systems based on biological models
02
using neural network models
04
Architecture, e.g. interconnection topology
G PHYSICS
06
COMPUTING; CALCULATING; COUNTING
N
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
7
Computer systems based on specific mathematical models
G PHYSICS
05
CONTROLLING; REGULATING
B
CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
13
Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
02
electric
04
involving the use of models or simulators
F MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
02
COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
P
IGNITION, OTHER THAN COMPRESSION IGNITION, FOR INTERNAL-COMBUSTION ENGINES; TESTING OF IGNITION TIMING IN COMPRESSION-IGNITION ENGINES
5
Advancing or retarding electric ignition spark; Control therefor
04
automatically, as a function of the working conditions of the engine or vehicle or of the atmospheric conditions
145
using electrical means
15
Digital data processing
G PHYSICS
06
COMPUTING; CALCULATING; COUNTING
F
ELECTRIC DIGITAL DATA PROCESSING
15
Digital computers in general; Data processing equipment in general
76
Architectures of general purpose stored programme computers
78
comprising a single central processing unit
Applicants: ROBERT BOSCH GMBH[DE/DE]; Postfach 30 02 20 70442 Stuttgart, DE
Inventors: GUNTORO, Andre; DE
Priority Data:
10 2016 216 944.007.09.2016DE
Title (EN) METHOD FOR CALCULATING A NEURON LAYER OF A MULTI-LAYER PERCEPTRON MODEL WITH SIMPLIFIED ACTIVATION FUNCTION
(FR) PROCÉDÉ POUR CALCULER UNE COUCHE NEURONALE D'UN MODÈLE DE PERCEPTRON MULTICOUCHE AU MOYEN D'UNE FONCTION D'ACTIVATION SIMPLIFIÉE
(DE) VERFAHREN ZUR BERECHNUNG EINER NEURONENSCHICHT EINES MEHRSCHICHTIGEN PERZEPTRONENMODELLS MIT VEREINFACHTER AKTIVIERUNGSFUNKTION
Abstract:
(EN) The invention relates to a method for calculating a neuron layer of a multi-layer perceptron model, comprising a hardwired calculation core (11, 13, 14) formed in a hardware for calculating a fixedly preset computing algorithm in coupled functional blocks, wherein a neuron (20) of a neuron layer of the perceptron model is calculated using an activation function, wherein the activation function corresponds to a simplified sigmoid function and a simplified tanh function, the activation function being formed by zero-point mirroring of the negative definition range of the exponential function.
(FR) L'invention concerne un procédé pour calculer une couche neuronale d'un modèle de perceptron multicouche comprenant un cœur de processeur (11, 13, 14) câblé agencé dans l'équipement informatique et conçu pour calculer un algorithme de calcul prédéterminé dans des blocs fonctionnels couplés, un neurone (20) de la couche neuronale du modèle de perceptron étant calculé au moyen d'une fonction d'activation, cette fonction d'activation correspondant à une fonction sigmoïde simplifiée et à une fonction tanh simplifiée, ladite fonction d'activation étant générée par une symétrie par rapport au point origine du domaine de définition négatif de la fonction exponentielle.
(DE) Die Erfindung betrifft ein Verfahren zur Berechnung einer Neuronenschicht eines mehrschichtigen Perzeptronenmodells mit einem in Hardware ausgebildeten fest verdrahteten Rechenkern (11, 13, 14) zur Berechnung eines fest vorgegebenen Rechenalgorithmus in gekoppelten Funktionsblöcken, wobei ein Neuron (20) einer Neuronenschicht des Perzeptronenmodells mithilfe einer Aktivierungsfunktion berechnet wird, wobei die Aktivierungsfunktion eine vereinfachte Sigmoidfunktion und eine vereinfachte tanh-Funktion entspricht, wobei die Aktivierungsfunktion durch Null-Punkt-Spiegelung des negativen Definitionsbereichs der Exponentialfunktion gebildet wird.
front page image
Designated States: AE, AG, AL, AM, AO, AT, AU, AZ, BA, BB, BG, BH, BN, BR, BW, BY, BZ, CA, CH, CL, CN, CO, CR, CU, CZ, DE, DJ, DK, DM, DO, DZ, EC, EE, EG, ES, FI, GB, GD, GE, GH, GM, GT, HN, HR, HU, ID, IL, IN, IR, IS, JO, JP, KE, KG, KH, KN, KP, KR, KW, KZ, LA, LC, LK, LR, LS, LU, LY, MA, MD, ME, MG, MK, MN, MW, MX, MY, MZ, NA, NG, NI, NO, NZ, OM, PA, PE, PG, PH, PL, PT, QA, RO, RS, RU, RW, SA, SC, SD, SE, SG, SK, SL, SM, ST, SV, SY, TH, TJ, TM, TN, TR, TT, TZ, UA, UG, US, UZ, VC, VN, ZA, ZM, ZW
African Regional Intellectual Property Organization (ARIPO) (BW, GH, GM, KE, LR, LS, MW, MZ, NA, RW, SD, SL, ST, SZ, TZ, UG, ZM, ZW)
Eurasian Patent Office (AM, AZ, BY, KG, KZ, RU, TJ, TM)
European Patent Office (EPO) (AL, AT, BE, BG, CH, CY, CZ, DE, DK, EE, ES, FI, FR, GB, GR, HR, HU, IE, IS, IT, LT, LU, LV, MC, MK, MT, NL, NO, PL, PT, RO, RS, SE, SI, SK, SM, TR)
African Intellectual Property Organization (BF, BJ, CF, CG, CI, CM, GA, GN, GQ, GW, KM, ML, MR, NE, SN, TD, TG)
Publication Language: German (DE)
Filing Language: German (DE)