Deep generative models have been widely used in several areas of NLP, and various techniques have been proposed to augment them or address their training challenges. In this paper, we propose a simple modification to Variational Autoencoders (VAEs) by using an Isotropic Gaussian Posterior (IGP) that allows for better utilisation of their latent representation space. This model avoids the sub-optimal behavior of VAEs related to inactive dimensions in the representation space. We provide both theoretical analysis, and empirical evidence on various datasets and tasks that show IGP leads to consistent improvement on several quantitative and qualitative grounds, from downstream task performance and sample efficiency to robustness. Additionally, we give insights about the representational properties encouraged by IGP and also show that its gain generalises to image domain as well.
翻译:深度基因模型已在国家实验室方案的几个领域广泛使用,并提出了各种技术来扩大这些模型,或应对其培训挑战。在本文件中,我们建议通过使用Isotrocic Gaussian Posteri (IGP) 来简单修改变式自动编码器(VAE),以便更好地利用其潜在代表空间。这一模型避免了VAEs与代表空间的不活动层面有关的次优行为。我们提供了关于各种数据集和任务的理论分析和经验证据,表明IGP在从下游任务性能和抽样效率到稳健性等若干方面导致持续改进的定量和定性理由。此外,我们还深入介绍了IGP鼓励的表示属性,并表明它获得了图像域的概括性。