Feedback
Settings

Settings

Goto Application

12,155 results
Offices all Languages en Stemming true Single Family Member false Include NPL false
RSS feed can only be generated if you have a WIPO account

Save query

A private query is only visible to you when you are logged-in and can not be used in RSS feeds

Query Tree

Refine Options

Offices
All
Specify the language of your search keywords
Stemming reduces inflected words to their stem or root form.
For example the words fishing, fished,fish, and fisher are reduced to the root word,fish,
so a search for fisher returns all the different variations
Returns only one member of a family of patents
Include Non-Patent literature in results

Full Query

AItechniqueMachineLearningRuleLearning

Side-by-side view shortcuts

General
Go to Search input
CTRL + SHIFT +
Go to Results (selected record)
CTRL + SHIFT +
Go to Detail (selected tab)
CTRL + SHIFT +
Go to Next page
CTRL +
Go to Previous page
CTRL +
Results (First, do 'Go to Results')
Go to Next record / image
/
Go to Previous record / image
/
Scroll Up
Page Up
Scroll Down
Page Down
Scroll to Top
CTRL + Home
Scroll to Bottom
CTRL + End
Detail (First, do 'Go to Detail')
Go to Next tab
Go to Previous tab

Analysis

1 / 1,216
1.20210397895INTELLIGENT LEARNING SYSTEM WITH NOISY LABEL DATA
US - 23.12.2021
Int.Class G06K 9/62
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
KGRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for recognising patterns
62Methods or arrangements for pattern recognition using electronic means
 Appl.No 16946465 Applicant INTERNATIONAL BUSINESS MACHINES CORPORATION Inventor Yang SUN

Various embodiments are provided for providing machine learning with noisy label data in a computing environment using one or more processors in a computing system. A label corruption probability of noisy labels may be estimated for selected data from a dataset using temporal inconsistency in a machine model prediction during a training operation in a neural network.

2.3786855AUTOMATED DATA PROCESSING AND MACHINE LEARNING MODEL GENERATION
EP - 03.03.2021
Int.Class G06N 5/00
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
5Computing arrangements using knowledge-based models
 Appl.No 19290076 Applicant ACCENTURE GLOBAL SOLUTIONS LTD Inventor HIGGINS LUKE
A device may obtain first data relating to a machine learning model. The device may pre-process the first data to alter the first data to generate second data. The device may process the second data to select a set of features from the second data. The device may analyze the set of features to evaluate a plurality of types of machine learning models with respect to the set of features. The device may select a particular type of machine learning model for the set of features based on analyzing the set of features to evaluate the plurality of types of machine learning models. The device may tune a set of parameters of the particular type of machine learning model to train the machine learning model. The device may receive third data for prediction. The device may provide a prediction using the particular type of machine learning model.
3.WO/2024/054286MACHINE LEARNING AND NATURAL LANGUAGE PROCESSING (NLP)-BASED SYSTEM FOR SYSTEM-ON-CHIP (SOC) TROUBLESHOOTING
WO - 14.03.2024
Int.Class G06F 30/33
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
30Computer-aided design
30Circuit design
32Circuit design at the digital level
33Design verification, e.g. functional simulation or model checking
 Appl.No PCT/US2023/026170 Applicant QUALCOMM INCORPORATED Inventor CAKIR, Murat
A method for processor-implemented method includes receiving an integrated circuit (IC) troubleshooting query for an IC (816). The IC troubleshooting query (816) is received from a user. The method also includes performing natural language processing and machine learning to cluster the IC troubleshooting query into one of a number of semantically similar troubleshooting categories. The method further includes retrieving resolution data from an expert system library (812), based on a mapping between categories of user solutions and a topic of the IC troubleshooting query. The method also includes generating a recommendation in response to the IC troubleshooting query, based on the resolution data (818). The method outputs the recommendation to the user.
4.20190056715Framework for rapid additive design with generative techniques
US - 21.02.2019
Int.Class G05B 19/4099
GPHYSICS
05CONTROLLING; REGULATING
BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
19Programme-control systems
02electric
18Numerical control , i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
4097characterised by using design data to control NC machines, e.g. CAD/CAM
4099Surface or curve machining, making 3D objects, e.g. desktop manufacturing
 Appl.No 15678653 Applicant General Electric Company Inventor Arun Karthi Subramaniyan

According to some embodiments, a system may include a design experience data store containing electronic records associated with prior industrial asset item designs. A deep learning model platform, coupled to the design experience data store, may include a communication port to receive constraint and load information from a designer device. The deep learning platform may further include a computer processor adapted to automatically and generatively create boundaries and geometries, using a deep learning model associated with an additive manufacturing process, for an industrial asset item based on the prior industrial asset item designs and the received constraint and load information. According to some embodiments, the deep learning model computer processor is further to receive design adjustments from the designer device. The received design adjustments might be for example, used to execute an optimization process and/or be fed back to continually re-train the deep learning model.

5.20030140020Plausible neural network with supervised and unsupervised cluster analysis
US - 24.07.2003
Int.Class G06E 1/00
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
EOPTICAL COMPUTING DEVICES
1Devices for processing exclusively digital data
 Appl.No 10294773 Applicant CHEN YUAN YAN Inventor Chen Yuan Yan

A plausible neural network (PLANN) is an artificial neural network with weight connection given by mutual information, which has the capability of inference and learning, and yet retains many characteristics of a biological neural network. The learning algorithm is based on statistical estimation, which is faster than the gradient decent approach currently used. The network after training becomes a fuzzy/belief network; the inference and weight are exchangeable, and as a result, knowledge extraction becomes simple. PLANN performs associative memory, supervised, semi-supervised, unsupervised learning and function/relation approximation in a single network architecture. This network architecture can easily be implemented by analog VLSI circuit design.

6.20180018757TRANSFORMING PROJECTION DATA IN TOMOGRAPHY BY MEANS OF MACHINE LEARNING
US - 18.01.2018
Int.Class G06T 3/40
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
3Geometric image transformations in the plane of the image
40Scaling of whole images or parts thereof, e.g. expanding or contracting
 Appl.No 15646119 Applicant Kenji SUZUKI Inventor Kenji SUZUKI

A method and system for transforming low-quality projection data into higher quality projection data, using of a machine learning model. Regions are extracted from an input projection image acquired, for example, at a reduced x-ray radiation dose (lower-dose), and pixel values in the region are entered into the machine learning model as input. The output of the machine learning model is a region that corresponds to the input region. The output information is arranged to form an output high-quality projection image. A reconstruction algorithm reconstructs high-quality tomographic images from the output high-quality projection images. The machine learning model is trained with matched pairs of projection images, namely, input lower-quality (lower-dose) projection images together with corresponding desired higher-quality (higher-dose) projection images. Through the training, the machine learning model learns to transform lower-quality (lower-dose) projection images to higher-quality (higher-dose) projection images. Once trained, the trained machine learning model does not require the higher-quality (higher-dose) projection images anymore. When a new lower-quality (low radiation dose) projection image is entered, the trained machine learning model would output a region similar to its desired region, in other words, it would output simulated high-quality (high-dose) projection images where noise and artifacts due to low radiation dose are substantially reduced, i.e., a higher image quality. The reconstruction algorithm reconstructs simulated high-quality (high-dose) tomographic images from the output high-quality (high-dose) projection images. With the simulated high-quality (high-dose) tomographic images, the detectability of lesions and clinically important findings can be improved.

7.4559381METHOD AND SYSTEM FOR TRAINING SELF-SUPERVISED LEARNING BASED-SLEEP STAGE CLASSIFICATION MODEL USING SMALL NUMBER OF LABELS
EP - 28.05.2025
Int.Class A61B 5/00
AHUMAN NECESSITIES
61MEDICAL OR VETERINARY SCIENCE; HYGIENE
BDIAGNOSIS; SURGERY; IDENTIFICATION
5Measuring for diagnostic purposes ; Identification of persons
 Appl.No 23843326 Applicant IUCF HYU Inventor CHAE DONG-KYU
Disclosed are a method and system for training a self-supervised learning based-sleep stage classification model using a small number of labels. The sleep stage classification method performed by a computer system according to an embodiment may comprising the steps of: receiving polysomnography data to be inputted into a self-supervised learning based-sleep stage classification model; and classifying sleep stages from the polysomnography data by using the self-supervised learning based-sleep stage classification model, wherein the self-supervised learning based-sleep stage classification model is trained on patterns for sleep stage classification from new sleep data through transfer learning by fine-tuning weights on the basis of a representation learning model that is trained on representations from sleep signal data.
8.20180341632Conversation utterance labeling
US - 29.11.2018
Int.Class G06F 17/24
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
17Digital computing or data processing equipment or methods, specially adapted for specific functions
20Handling natural language data
21Text processing
24Editing, e.g. insert/delete
 Appl.No 15603091 Applicant International Business Machines Corporation Inventor Rama Kalyani T. Akkiraju

A method, a computer program product, and an information handling system is provided for labeling unlabeled utterances given a taxonomy of labels utilizing topic word semi-supervised learning.

9.20200111005Trusted neural network system
US - 09.04.2020
Int.Class G06N 3/08
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computing arrangements based on biological models
02Neural networks
08Learning methods
 Appl.No 16226286 Applicant SRI International Inventor Shalini Ghosh

In general, the disclosure describes techniques for facilitating trust in neural networks using a trusted neural network system. For example, described herein are multi-headed, trusted neural network systems that can be trained to satisfy one or more constraints as part of the training process, where such constraints may take the form of one or more logical rules and cause the objective function of at least one the heads of the trusted neural network system to steer, during machine learning model training, the overall objective function for the system toward an optimal solution that satisfies the constraints. The constraints may be non-temporal, temporal, or a combination of non-temporal and temporal. The constraints may be directly compiled to a neural network or otherwise used to train the machine learning model.

10.4134879METHOD AND SYSTEM FOR TRAINING A MACHINE LEARNING MODEL
EP - 15.02.2023
Int.Class G06N 5/02
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
5Computing arrangements using knowledge-based models
02Knowledge representation; Symbolic representation
 Appl.No 21190562 Applicant SIEMENS AG Inventor HILDEBRANDT MARCEL
An initially trained machine learning model is used by an active learning module to generate candidate triples, which are fed into an expert system for verification. As a result, the expert system outputs novel facts that are used for retraining the machine learning model. This approach consolidates expert systems with machine learning through iterations of an active learning loop, by bringing the two paradigms together, which is in general difficult because training of a neural network (machine learning) requires differentiable functions and rules (used by expert systems) tend not to be differentiable. The method and system provide a data augmentation strategy where the expert system acts as an oracle and outputs the novel facts, which provide labels for the candidate triples. The novel facts provide critical information from the oracle that is injected into the machine learning model at the retraining stage, thus allowing to increase its generalization performance. Furthermore, deployment effort and maintenance costs are reduced, since in industrial AI applications mentioned above, only the machine learning model needs to be deployed instead of deploying a machine learning model as well as an expert system. According to an embodiment, the active learning module identifies candidate triples, for which, if the truth value was known, it would maximally boost the performance of the machine learning model. In other words, the active learning module chooses candidate triples representing unknown facts that lie close to the decision boundary of the machine learning model. In this way, the novel facts are sampled efficiently from the expert system such that the generalization performance of the machine learning model is maximized.
# -