Naver. has filed a patent for a sampling system that utilizes an energy-based model to generate scores for input classifications. The system includes a sampling module that generates samples based on a probability distribution, with a mechanism to determine acceptance probability and discard samples accordingly. GlobalData’s report on Naver gives a 360-degree view of the company including its patenting strategy. Buy the report here.

According to GlobalData’s company profile on Naver, Recommender systems was a key innovation area identified from patents. Naver's grant share as of January 2024 was 57%. Grant share is based on the ratio of number of grants to total number of patents.

Sampling system using energy-based model for generating non-negative scores

Source: United States Patent and Trademark Office (USPTO). Credit: Naver Corp

A patent application (Publication Number: US20240037184A1) describes a sampling system that includes an energy-based model (EBM) generating non-negative scores for input with discrete classifications. The system also features a sampling module that generates samples from a probability distribution of the EBM using a proposal distribution. The module sets a probability of acceptance for the sample based on various factors, including a minimum value and an acceptance value determined from the sample's score, the proposal distribution, and an upper boundary value. The system further involves determining a distribution value between 0 and 1 using a uniform distribution and discarding the sample if the distribution value exceeds the probability of acceptance.

Moreover, the sampling system can adjust parameters such as the upper boundary value based on differences in probabilities or KL divergence to maintain an acceptance rate of samples above a predetermined minimum. Additionally, the patent application extends to a question answering system incorporating the sampling system, where the input is a question, and the system outputs the sample as the answer when the distribution value is below the probability of acceptance. The patent also covers a sampling method mirroring the system's functionality, including generating samples from the EBM's probability distribution, setting acceptance probabilities, and adjusting parameters to ensure a minimum acceptance rate is maintained.

To know more about GlobalData’s detailed insights on Naver, buy the report here.

Premium 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

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.