Variational Autoencoders (VAEs) are powerful generative models that have been widely used in various fields, including image and text generation. However, one of the known challenges in using VAEs is the model's sensitivity to its hyperparameters, such as the latent space size. This paper presents a simple extension of VAEs for automatically determining the optimal latent space size during the training process by gradually decreasing the latent size through neuron removal and observing the model performance. The proposed method is compared to traditional hyperparameter grid search and is shown to be significantly faster while still achieving the best optimal dimensionality on four image datasets. Furthermore, we show that the final performance of our method is comparable to training on the optimal latent size from scratch, and might thus serve as a convenient substitute.

翻译：变分自编码器（VAE）是一种强大的生成模型，已被广泛应用于图像和文本生成等领域。然而，使用VAE时的一个已知挑战是其对超参数（如隐空间大小）的敏感性。本文提出了一种VAE的简单扩展方法，通过逐步移除神经元并观察模型性能，在训练过程中自动确定最优的隐空间大小。所提方法与传统的超参数网格搜索方法进行了比较，结果表明，该方法在四个图像数据集上显著更快，同时仍能实现最优维度。此外，我们证明该方法最终性能与从头开始使用最优隐空间大小训练的结果相当，因此可作为一种便捷的替代方案。