首页|Osaka University Researchers Yield New Data on Machine Learning (Fast, accurate, and interpretable decoding of electrocorticographic signals using dynamic mode decomposition)
Osaka University Researchers Yield New Data on Machine Learning (Fast, accurate, and interpretable decoding of electrocorticographic signals using dynamic mode decomposition)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – New research on artificial intelligenc e is the subject of a new report. According to news reporting originating from O saka University by NewsRx correspondents, research stated, “Dynamic mode (DM) de composition decomposes spatiotemporal signals into basic oscillatory components (DMs).” Funders for this research include Mext | Japan Science And Technology Agency; Ja pan Agency For Medical Research And Development; Mext | Japan Society For The Pr omotion of Science. Our news correspondents obtained a quote from the research from Osaka University : “DMs can improve the accuracy of neural decoding when used with the nonlinear Grassmann kernel, compared to conventional power features. However, such kernel- based machine learning algorithms have three limitations: large computational ti me preventing real-time application, incompatibility with non-kernel algorithms, and low interpretability. Here, we propose a mapping function corresponding to the Grassmann kernel that explicitly transforms DMs into spatial DM (sDM) featur es, which can be used in any machine learning algorithm. Using electrocorticogra phic signals recorded during various movement and visual perception tasks, the s DM features were shown to improve the decoding accuracy and computational time c ompared to conventional methods. Furthermore, the components of the sDM features informative for decoding showed similar characteristics to the high-g power of the signals, but with higher trial-to-trial reproducibility.”
NETL
NSTL
