Ossi Räisä, Stratis Markou, Matthew Ashman, Wessel P. Bruinsma, Marlon Tobaben...
TL;DR通过使用模拟数据来训练元学习模型，将Convolutional Conditional Neural Process (ConvCNP) 与改进的DP机制相结合，从而提供准确、良好校准的预测模型，并在非高斯数据上优于DP高斯过程（GP）基线模型。
Abstract
Many high-stakes applications require machine learning models that protect user privacy and provide well-calibrated, accurate predictions. While differential privacy (DP) is the gold standard for protecting user