Semi-Supervised Learning via Compact Latent Space Clustering
Semi-Supervised Learning via Compact Latent Space Clustering
复制标题
DOI：
--
复制
发表时间：
2018-06
期刊：
影响因子：
--
通讯作者：
K. Kamnitsas;Daniel Coelho de Castro;L. L. Folgoc-L.;Ian Walker;Ryutaro Tanno;D. Rueckert;Ben Glocker;A. Criminisi;A. Nori
中科院分区：
文献类型：
--
作者：
K. Kamnitsas;Daniel Coelho de Castro;L. L. Folgoc-L.;Ian Walker;Ryutaro Tanno;D. Rueckert;Ben Glocker;A. Criminisi;A. Nori
我们为神经网络的半监督学习提出了一种新颖的成本函数，它鼓励潜在空间的紧凑聚类以促进分离。关键思想是在训练批次的标记和未标记样本的嵌入上动态创建图，以捕获特征空间中的底层结构，并使用标签传播来估计其高密度和低密度区域。然后，我们设计了一个基于图上马尔可夫链的成本函数，该函数对潜在空间进行正则化，以形成每个类的单个紧凑簇，同时避免在优化过程中干扰现有簇。我们根据三个基准评估我们的方法，并与最先进的方法进行比较，结果有希望。我们的方法结合了基于图的正则化和高效归纳推理的优点，不需要修改网络架构，因此可以轻松应用于现有网络，以实现未标记数据的有效使用。
We present a novel cost function for semi-supervised learning of neural networks that encourages compact clustering of the latent space to facilitate separation. The key idea is to dynamically create a graph over embeddings of labeled and unlabeled samples of a training batch to capture underlying structure in feature space, and use label propagation to estimate its high and low density regions. We then devise a cost function based on Markov chains on the graph that regularizes the latent space to form a single compact cluster per class, while avoiding to disturb existing clusters during optimization. We evaluate our approach on three benchmarks and compare to state-of-the art with promising results. Our approach combines the benefits of graph-based regularization with efficient, inductive inference, does not require modifications to a network architecture, and can thus be easily applied to existing networks to enable an effective use of unlabeled data.