Authors
- Yuhta Takida
- Takashi Shibuya
- Wei-Hsiang Liao
- Chieh-Hsin Lai
- Junki Ohmura*
- Toshimitsu Uesaka
- Naoki Murata
- Shusuke Takahashi*
- Toshiyuki Kumakura*
- Yuki Mitsufuji
* External authors
Venue
- ICML 2022
Date
- 2022
Share
SQ-VAE: Variational Bayes on Discrete Representation with Self-annealed Stochastic Quantization
Junki Ohmura*
Shusuke Takahashi*
Toshiyuki Kumakura*
* External authors
ICML 2022
2022
Abstract
One noted issue of vector-quantized variational autoencoder (VQ-VAE) is that the learned discrete representation uses only a fraction of the full capacity of the codebook, also known as codebook collapse. We hypothesize that the training scheme of VQ-VAE, which involves some carefully designed heuristics, underlies this issue. In this paper, we propose a new training scheme that extends the standard VAE via novel stochastic dequantization and quantization, called stochastically quantized variational autoencoder (SQ-VAE). In SQ-VAE, we observe a trend that the quantization is stochastic at the initial stage of the training but gradually converges toward a deterministic quantization, which we call self-annealing. Our experiments show that SQ-VAE improves codebook utilization without using common heuristics. Furthermore, we empirically show that SQ-VAE is superior to VAE and VQ-VAE in vision- and speech-related tasks.
Related Publications
Classifier-Free Guidance (CFG) is a widely used technique for conditional generation and improving sample quality in continuous diffusion models, and recent works have extended it to discrete diffusion. This paper theoretically analyzes CFG in the context of masked discrete …
We present 3DScenePrompt, a framework that generates the next video chunk from arbitrary-length input while enabling precise camera control and preserving scene consistency. Unlike methods conditioned on a single image or a short clip, we employ dual spatio-temporal conditio…
This paper introduces LLM2Fx-Tools, a multimodal tool-calling framework that generates executable sequences of audio effects (Fx-chain) for music post-production. LLM2Fx-Tools uses a large language model (LLM) to understand audio inputs, select audio effects types, determine…
JOIN US
Shape the Future of AI with Sony AI
We want to hear from those of you who have a strong desire
to shape the future of AI.
