论文标题
投影性潜在干预措施，用于理解和微调分类器
Projective Latent Interventions for Understanding and Fine-tuning Classifiers
论文作者
论文摘要
众所周知，由神经网络分类器学到的高维度潜在表示很难解释。尤其是在医疗应用中，模型开发人员和领域专家希望更好地了解这些潜在表示与所得分类性能如何相关的。我们提出了投影潜在干预措施（PLIS），这是一种通过反向传播对潜在空间低维嵌入的手动更改来重新培训分类器的技术。后传播基于T分布的随机邻嵌入的参数近似。 PLIS允许域专家以直观的方式控制潜在决策空间，以更好地符合他们的期望。例如，可以通过手动将嵌入中的类群体分开来增强特定的类别的性能。我们在胎儿超声成像中的真实情况下评估了我们的技术。
High-dimensional latent representations learned by neural network classifiers are notoriously hard to interpret. Especially in medical applications, model developers and domain experts desire a better understanding of how these latent representations relate to the resulting classification performance. We present Projective Latent Interventions (PLIs), a technique for retraining classifiers by back-propagating manual changes made to low-dimensional embeddings of the latent space. The back-propagation is based on parametric approximations of t-distributed stochastic neighbourhood embeddings. PLIs allow domain experts to control the latent decision space in an intuitive way in order to better match their expectations. For instance, the performance for specific pairs of classes can be enhanced by manually separating the class clusters in the embedding. We evaluate our technique on a real-world scenario in fetal ultrasound imaging.
