首页|'First-To-Saturate Single Modal Latent Feature Activation For Explanation Of Mac hine Learning Models' in Patent Application Approval Process (USPTO 20240202516)
'First-To-Saturate Single Modal Latent Feature Activation For Explanation Of Mac hine Learning Models' in Patent Application Approval Process (USPTO 20240202516)
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
The following quote was obtained by the news editors from the background informa tion supplied by theinventors: “Machine learning models, such as neural network s, may be used in critical applications suchas in the healthcare, manufacturing , transportation, financial, information technology industries, amongothers. In these and other applications, explanations to a user related to why the model g enerated aspecific prediction for a particular input, what data, models, and pr ocessing have been applied to generatethat prediction, and/or the like can be u seful, and in some instances, required. However, conventionalexplainable machin e learning methods inefficiently and/or inaccurately demonstrate the properties of thehidden units of the models as they fail to address the multi-modal nature of unconstrained latent featureactivation. As a result, explainability methods on conventional models provide inconsistent and unreliableexplanation associat ed with the model output.”
NETL
NSTL
