摘要
本专利申请未转让给公司或机构。以下引文是新闻编辑从新闻编辑提供的背景资料中获得的发明者：“诸如人工现实耳机、sm艺术眼镜、智能手表等设备，以及其他用户用于做出各种类型预测的机器学习模型。机器学习模型可以是用于图像中的目标识别、语音识别等。例如，设备可以响应可以使用机器学习模型来识别唤醒单词。这些机器学习模型可能有大量的参数，例如几百万个参数。由于物理尺寸的限制，这种设备具有小的计算能力和小的存储数据的存储器。作为结果，这些设备可能无法存储大型机器学习模型或多个机器学习处理不同类型信息的模型。
Abstract
This patent application has not been assigned to a company or institution.The following quote was obtained by the news editors from the background informa tion supplied bythe inventors: “Devices such as artificial reality headsets, sm art glasses, smart watches, and others usermachine learning models for making v arious types of predictions. The machine learning models may beused for object recognition in images, for speech recognition, and so on. For example, a device may respondto a wake word and a machine learning model may be used for recogniz ing the wake word. These machinelearning models may have large number of parame ters, for example, several million parameters. Due tophysical size limitations, such devices have small computing power and small memory for storing data. Asa result, such devices may not be able to store large machine learning models or multiple machine learningmodels for processing different type of information.
NETL
NSTL