A system (100) is disclosed for training a generative model and a discriminative model. The generative model generates synthetic instances from latent feature vectors by generating an intermediate representation from the latent feature vector and generating the synthetic instance from the intermediate representation. The discriminative model determines multiple discriminator scores for multiple parts of an input instance, indicating whether the part is from a synthetic instance or an actual instance. The generative model is trained by backpropagation. During the backpropagation, partial derivatives of the loss with respect to entries of the intermediate representation are updated based on a discriminator score for a part of the synthetic instance, wherein the part of the synthetic instanceis generated based at least in part on the entry of the intermediate representation, and wherein the partial derivative is decreased in value if the discriminator score indicates an actual instance.
公开了一种用于训练生成模型和判别模型的系统(100)。生成模型通过从潜在特征向量生成中间表示并从中间表示生成合成实例来从潜在特征向量生成合成实例。判别模型确定用于输入实例的多个部分的多个判别器得分,其指示部分是来自合成实例还是实际实例。通过反向传播来训练生成模型。在反向传播期间,基于用于合成实例的部分的判别器得分来更新损失相对于中间表示的条目的偏导数,其中至少部分基于中间表示的条目来生成合成实例的部分,并且其中如果判别器得分指示实际实例,则偏导数的值减小。
TRAINING A GENERATIVE MODEL AND A DISCRIMINATIVE MODEL
训练生成模型和判别模型
2021-01-19
Patent
Elektronische Ressource
Chinesisch
IPC: | G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / B60R Fahrzeuge, Fahrzeugausstattung oder Fahrzeugteile, soweit nicht anderweitig vorgesehen , VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR / G05D SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES , Systeme zum Steuern oder Regeln nichtelektrischer veränderlicher Größen |
Hybrid Generative-Discriminative Visual Categorization
British Library Online Contents | 2008
|British Library Online Contents | 2015
|A Generative/Discriminative Learning Algorithm for Image Classification
British Library Conference Proceedings | 2005
|