首页|Researchers Submit Patent Application, 'Identifying Performance Degradation In M achine Learning Models Based On Comparison Of Actual And Predicted Results', for Approval (USPTO 20240112010)
Researchers Submit Patent Application, 'Identifying Performance Degradation In M achine Learning Models Based On Comparison Of Actual And Predicted Results', for Approval (USPTO 20240112010)
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
News editors obtained the following quote from the background information suppli ed by the inventors:“In recent years, the use of artificial intelligence, inclu ding, but not limited to, machine learning, deeplearning, etc. (referred to col lectively herein as artificial intelligence models, machine learning models,or simply models) has exponentially increased. Broadly described, artificial intell igence refers to a widerangingbranch of computer science concerned with buildi ng smart machines capable of performing tasksthat typically require human intel ligence. Key benefits of artificial intelligence are its ability to processdata , find underlying patterns, and/or perform real-time determinations. However, de spite these benefitsand despite the wide-ranging number of potential applicatio ns, practical implementations of artificialintelligence have been hindered by s everal technical problems. Results based on artificial intelligence arenotoriou sly difficult to review as the process by which the results are made may be unkn own or obscured.This obscurity creates hurdles for identifying errors in the re sults and for improving the models providingthose results. For example, perform ance of machine learning models may degrade over time. Generally, auser is able to determine that a machine learning model has degraded based on comparing actu al resultsto predicted results. However, some datasets include values predicted based on data from multiple machinelearning models. For example, a particular value may be calculated from a combination of values that arederived from diffe rent machine learning models. Thus, it is difficult to determine which machine l earningmodel’s performance has degraded.”
Artificial IntelligenceBusinessCapit al One Services LLCCyborgsEmerging TechnologiesMachine Learning
NETL
NSTL
