InfoVAE: Information Maximizing Variational Autoencoders
This addresses a fundamental limitation in variational autoencoder training for researchers in generative modeling.
The paper tackles the problem of variational autoencoders producing inaccurate amortized inference distributions and ignoring latent variables with flexible decoders, proposing InfoVAE objectives that significantly improve variational posterior quality and latent feature utilization.
A key advance in learning generative models is the use of amortized inference distributions that are jointly trained with the models. We find that existing training objectives for variational autoencoders can lead to inaccurate amortized inference distributions and, in some cases, improving the objective provably degrades the inference quality. In addition, it has been observed that variational autoencoders tend to ignore the latent variables when combined with a decoding distribution that is too flexible. We again identify the cause in existing training criteria and propose a new class of objectives (InfoVAE) that mitigate these problems. We show that our model can significantly improve the quality of the variational posterior and can make effective use of the latent features regardless of the flexibility of the decoding distribution. Through extensive qualitative and quantitative analyses, we demonstrate that our models outperform competing approaches on multiple performance metrics.