Trustworthy Deep Learning for Large-scale Online Advertisement

Project: Research project

Project Details

Description

本研究旨在研究复杂弱监督情况下的鲁棒机器学习,设计能够在真实场景下应对噪声行为的通用鲁棒算法、网络模型、训练框架。针对噪声行为在大外投业务中的具体表现,我们将其建模为以下四个主要问题,即流量稳定性差,外部用户行为稀疏,上下文信息缺失,行为预测带噪。对此,本研究提出四个相应的任务尝试解决对应的技术难点,并进行业务场景的适配。
StatusFinished
Effective start/end date21/06/2221/06/23

Fingerprint

Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.