This post is also available in: עברית (Hebrew)
A new method of continuous learning of data for artificial intelligence has been developed by a research group from the Osaka Metropolitan University. This new method combines classification performance for data with multiple labels with the ability to learn continually from data.
Numerical experiments on real-world multi-label data indicate that the new method outperforms conventional approaches. The simplicity of this algorithm makes it easy to integrate it with other algorithms to devise new ones, according to sciencedaily.com.
As we continue to obtain more and more data while the world advancing towards IoT technology and smart living, artificial intelligence will no doubt play a huge part as a major tool to mitigate significant amounts of data.
To answer the growing demand, a research group led by Associate Professor Naoki Masuyama and Professor Yusuke Nojima of the Osaka Metropolitan University, has developed a new method that combines classification performance for data with multiple labels, with the ability to continually learn with data.
The simplicity of this new algorithm makes it easy to devise an evolved version which can be integrated with other algorithms. Since the underlying clustering method groups data based on the similarity between data entries, it is expected to be a useful tool for continual big data preprocessing by learning the data and learning the label information corresponding to the data separately and continually, so that both high classification performance and continual learning capability are achieved.
Prepared to dive into the world of futuristic technology? Attend INNOTECH 2023, the international convention and exhibition for cyber, HLS and innovation at Expo, Tel Aviv, on March 29th-30th
Interested in sponsoring / a display booth at the 2023 INNOTECH exhibition? Click here for details!