Chinese Academy of Sciences

Boe Technology Group Co., Ltd.

Hefei Xinsheng Optoelectronics Technology Co., Ltd.

Beijing Baidu Netcom Science and Technology Co., Ltd.

Yunlong Wang

Qi Cheng Li

Yong He

Zhenan Sun

Kunbo Zhang

Zhengquan Luo

Tao Ma

Chengshao Yang

Yu Ji

Yuming Li

Shengli Liu

Wenxiang Lu

Disclosed is a disentangled personalized federated learning method via consensus representation extraction and diversity propagation provided by embodiments of the present application. The method includes: receiving, by a current node, local consensus representation extraction models and unique representation extraction models corresponding to other nodes, respectively; extracting, by the current node, the representations of the data of the current node by using the unique representation extraction models of other nodes respectively, and calculating first mutual information between different sets of representation distributions, determining similarity of the data distributions between the nodes based on the size of the first mutual information, and determining aggregation weights corresponding to the other nodes based on the first mutual information; the current node obtains the global consensus representation aggregation model corresponding to the current node.

Disentangled personalized federated learning method via consensus representation extraction and diversity propagation

Disclosed is a method for updating a node model that resists discrimination propagation in federated learning. The method includes: obtaining a node model corresponding to a data node; calculating a mean value of the distribution of class features and a quantity ratio corresponding to training data of the data node, calculating a distribution weighted aggregation model based on the node model, the mean value of the distribution of class features and the quantity ratio; calculating a regularization term corresponding to the data node based on the node model and the distribution weighted aggregation model; calculating a variance of the distribution of the class features corresponding to the data node, calculating a class balanced complementary term by using a cross-domain feature generator; and updating the node model based on the distribution weighted aggregation model, the regularization term, and the class balanced complementary term.

Method for updating a node model that resists discrimination propagation in federated learning

The present disclosure provides a display substrate, a manufacturing method thereof, and a display device. The display substrate includes a base substrate and thin-film transistors disposed on the base substrate, where the thin-film transistors each comprises a gate, an active layer insulated from the gate, and two ohmic contact parts in direct contact with the active layer, leaving a gap region between the two ohmic contact parts; and each of the ohmic contact parts comprises a lightly doped region and a heavily doped region, and an orthographic projection of the lightly doped region on the base substrate and an orthographic projection of the heavily doped region on the base substrate do not overlap each other.

Display substrate, manufacturing method thereof, and display device

Embodiments of the present disclosure relate to a method and apparatus for assigning a computing task. The method includes: segmenting, based on a computing resource required for a computing task of a graph, the graph into a subgraph set; determining, based on a computing resource required for a computing task of a subgraph in the subgraph set, whether a node in a node cluster is adapted to the subgraph in the subgraph set; assigning, in response to determining the node in the node cluster being adapted to the subgraph in the subgraph set, the computing task of the subgraph in the subgraph set to the node adapted to the subgraph in the subgraph set, to compute the subgraph; and synthesizing a computing result of the graph based on a computing result of each subgraph in the subgraph set.

Method and apparatus for assigning computing task

Growing community of inventors

Beijing, China

Yunlong Wang