Huawei Technologies Co., Limited

International Business Machines Corporation

Beijing Sensetime Technology Development Co., Ltd.

Zhao Zhong

Shuang Shuang Li

Junjie Yan

Yun Ting Wang

Guang Ming Zhang

Pu Zhu

Chenglin Liu

Hai Lei Song

Yucong Zhou

Yikang Zhang

This application relates to the field of artificial intelligence technologies, and describes a classification model training method, a hyperparameter search method, and an apparatus. The training method includes obtaining a target hyperparameter of a to-be-trained classification model. The target hyperparameter is used to control a gradient update operation of the to-be-trained classification model. The to-be-trained classification model includes a scaling invariance linear layer. The scaling invariance linear layer enables a predicted classification result output when a weight parameter of the to-be-trained classification model is multiplied by any scaling coefficient to remain unchanged. The method further includes updating the weight parameter of the to-be-trained classification model based on the target hyperparameter and a target training manner, to obtain a trained classification model.

Classification model training method, hyperparameter search method, and apparatus

This application discloses a neural network construction method and apparatus, and an image processing method and apparatus in the field of artificial intelligence. The neural network construction method includes: constructing a search space based on an application requirement of a target neural network, where the search space includes M elements, the M elements are used to indicate M network structures, each of the M elements includes a quantity of blocks in a stage in a corresponding network structure and a channel quantity of each block, and M is a positive integer; and selecting a target network structure from the M network structures based on a distribution relationship among unevaluated elements in the search space. According to the method, a neural network satisfying a performance requirement can be efficiently constructed.

Neural network construction method and apparatus, and image processing method and apparatus

Embodiments of the present application disclose a method and apparatus for generating a neural network structure, an electronic device, and a storage medium. The method comprises: sampling a neural network structure to generate a network block, the network block comprising at least one network layer; constructing a sampling neural network based on the network block; training the sampling neural network based on sample data, and obtaining an accuracy corresponding to the sampling neural network; and in response to that the accuracy does not meet a preset condition, regenerating a new network block according to the accuracy until a sampling neural network constructed by the new network block meets the preset condition, and using the sampling neural network meeting the preset condition as a target neural network.

Method and apparatus for generating target neural network structure, electronic device, and storage medium

As disclosed herein a method, executed by a computer, includes monitoring proximate automobiles using a camera, receiving a request to transmit a communication connection request to a selected automobile, and determining observed attributes corresponding to the selected automobile based on images from the camera. The method further includes broadcasting, over a network, the observed attributes to the proximate automobiles, and requesting disclosed attributes and a connection identifier from the proximate automobiles that match the observed attributes, receiving at least one response from the proximate automobiles that match the observed attributes, and determining which response is a best match to the selected automobile. The method further includes transmitting the communication connection request to the selected automobile over the network using the connection identifier corresponding to the best match. A computer program product corresponding to the above method is also disclosed herein.

Enhanced communication system

Growing community of inventors

Beijing, China

Zhao Zhong