One-Shot Learning in Discriminative Neural Networks
One-Shot Learning in Discriminative Neural Networks
复制标题
DOI：
10.17863/cam.12051
复制
发表时间：
2017-07
期刊：
影响因子：
--
通讯作者：
Jordan Burgess;J. Lloyd;Zoubin Ghahramani
中科院分区：
文献类型：
--
作者：
Jordan Burgess;J. Lloyd;Zoubin Ghahramani
我们考虑的任务一次性学习的视觉类别。在本文中，我们探讨了贝叶斯过程更新预训练的convnet分类一个新的图像类别的数据是有限的。我们将这个convnet分解为固定的特征提取器和softmax分类器。我们假设新任务的目标权重来自与预训练的softmax权重相同的分布，我们将其建模为多变量高斯分布。通过将其用作新权重的先验，我们展示了与最先进方法的竞争性能，同时也与在大数据上训练深度网络的“正常”方法保持一致。
We consider the task of one-shot learning of visual categories. In this paper we explore a Bayesian procedure for updating a pretrained convnet to classify a novel image category for which data is limited. We decompose this convnet into a fixed feature extractor and softmax classifier. We assume that the target weights for the new task come from the same distribution as the pretrained softmax weights, which we model as a multivariate Gaussian. By using this as a prior for the new weights, we demonstrate competitive performance with state-of-the-art methods whilst also being consistent with 'normal' methods for training deep networks on large data.