| 发明名称 | Hierarchical hybrid batch-incremental learning | ||
| 摘要 | In one embodiment, a machine learning model for predicting one or more metrics is run in a network which includes a centralized controller device interconnected with a plurality of edge devices. A batch version of the machine learning model that operates in batch mode is hosted at the centralized controller device. Then, an incremental version of the machine learning model that operates in incremental mode is pushed to an edge device of the plurality of edge devices, such that the incremental version of the machine learning model is hosted at the edge device. As a result, the batch version and the incremental version of the machine learning model run in parallel with one another. | ||
| 申请公布号 | US9547828(B2) | 申请公布日期 | 2017.01.17 |
| 申请号 | US201414120371 | 申请日期 | 2014.05.14 |
| 申请人 | Cisco Technology, Inc. | 发明人 | Mermoud Grégory;Vasseur Jean-Philippe;Bouchacourt Diane |
| 分类号 | G06F17/00;G06N5/02;G06N99/00;G06N5/04;H04L29/08;H04L12/70;H04L12/26;H04L12/721;H04L12/751 | 主分类号 | G06F17/00 |
| 代理机构 | Parker Ibrahim & Berg LLC | 代理人 | Parker Ibrahim & Berg LLC ;Behmke James M.;LeBarron Stephen D. |
| 主权项 | 1. A method, comprising: running a machine learning model for predicting one or more metrics in a network which includes a centralized controller device interconnected with a plurality of edge devices; hosting, at the centralized controller device, a batch version of the machine learning model that operates in batch mode; and pushing an incremental version of the machine learning model that operates in incremental mode to an edge device of the plurality of edge devices, such that the incremental version of the machine learning model is hosted at the edge device, wherein the batch version and the incremental version of the machine learning model run in parallel with one another. | ||
| 地址 | San Jose CA US | ||