Samsung SDS had 20 patents in artificial intelligence during Q1 2024. The patents filed by Samsung SDS Co Ltd in Q1 2024 focus on methods and systems for domain adaptation, evaluating performance, semantic segmentation, embedding transformation, and embedding data. These inventions involve techniques such as generating pseudo labels, extracting support features, and transforming embedding vectors to improve the efficiency and accuracy of machine learning models. GlobalData’s report on Samsung SDS gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
Samsung SDS grant share with artificial intelligence as a theme is 20% in Q1 2024. Grant share is based on the ratio of number of grants to total number of patents.
Recent Patents
Application: Method for domain adaptation and system thereof (Patent ID: US20240086765A1)
The patent filed by Samsung SDS Co Ltd. describes a method and system for domain adaptation. The method involves obtaining a model trained on a labeled dataset from a source domain, selecting a data sample from an unlabeled dataset in a target domain, generating a second data sample through data augmentation, calculating a consistency loss between the two samples, and updating the feature extractor based on this loss. The system includes processors and memory to execute instructions for obtaining the model, selecting data samples, calculating consistency loss, and updating the feature extractor.
The method further includes training the model using only the unlabeled dataset of the target domain, calculating consistency loss based on differences in features extracted from data samples, generating pseudo labels for data samples, predicting labels, and updating the feature extractor based on entropy loss. Additionally, the system allows for updating the feature extractor while keeping the predictor frozen, and involves additional training on the unlabeled dataset of the target domain. The method and system aim to improve domain adaptation by efficiently utilizing labeled and unlabeled data from different domains to enhance the model's performance.
To know more about GlobalData’s detailed insights on Samsung SDS, buy the report here.
Data Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.
GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.
GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.
