Differentially Private Model Publishing For Deep Learning

Differentially Private Model Publishing For Deep Learning Calton Pu Presented at the 2019 IEEE Symposium on Security Privacy May 2022, 2019 San Francisco, CA Deep learning techniques based on neural networks have shown significant success in a wide range of AI tasks. Largescale training datasets are one of the critical factors for their success. However, when the training datasets are crowdsourced from individuals and contain sensitive information,