Akira Nakagawa — accepted AI-conference papers

2021

Quantitative Understanding of VAE as a Non-linearly Scaled Isometric Embedding

Akira Nakagawa, Keizo Kato, Taiji Suzuki

ICML 2021spotlight

Variational autoencoder (VAE) estimates the posterior parameters (mean and variance) of latent variables corresponding to each input data. While it is used for many tasks, the transparency of the model is still an underlying issue. This paper provides a quantitative understanding of VAE property thr…

Cited by 12 Source ↗PDF ↗Scholar ↗

2020

Rate-distortion optimization guided autoencoder for isometric embedding in Euclidean latent space

Keizo Kato, Jing Zhou, Tomotake Sasaki, Akira Nakagawa

ICML 2020poster

To analyze high-dimensional and complex data in the real world, deep generative models, such as variational autoencoder (VAE) embed data in a low-dimensional space (latent space) and learn a probabilistic model in the latent space. However, they struggle to accurately reproduce the probability distr…

Cited by 26 Source ↗PDF ↗Scholar ↗