2results about How to "Reduce system overhead" patented technology

This application discloses a data access method and apparatus, and a first computing device, and belongs to the field of computer technologies. The data access method includes: The first computing device receives a second write operation command sent by a second computing device, where the second write operation command is obtained after the second computing device preprocesses a first write operation command generated by a host, and the first write operation command is used to write to-be-written data into a storage pool; the first computing device performs a data processing operation on the to-be-written data according to the second write operation command; and the first computing device writes processed to-be-written data into the storage pool. This application reduces system overheads of the host.

The application discloses a kind of federated meta-learning image classification methods based on sparse neural network, communication efficiency is high, algorithm performance is good.The method of the present application mainly includes the following steps: (10) set up fully connected neural network node and correlation coefficient, establish source node set and target node set;(20) the full connection neural network is sparsified, and the weight parameters of sparse neural network are initialized;(30) training internal update is carried out on the training set data of each source node;External update is carried out on the test set data of each source node;(40) remove the minimum positive number and the maximum negative number of each layer of weight matrix after external update of each source node, whether iteration number t is the integer multiple of the set local iteration number is judged to make corresponding processing;(50) gradient update training is carried out on the training set data in each target node based on the parameters obtained by training source node set, and image classification task is carried out using the fine-tuned model of target node.