LL-VQ-VAE: Learnable Lattice Vector-Quantization For Efficient Representations

Ahmed Khalil, Robert Piechocki, Raul Santos-RodriguezSubmission to ICLR 2024 · Rejected

Abstract

In this paper we introduce $\text{\textit{learnable lattice}}$ vector quantization and demonstrate its effectiveness for learning discrete representations. Our method, termed LL-VQ-VAE, replaces the vector quantization layer in VQ-VAE with lattice-based discretization. The learnable lattice imposes a structure over all discrete embeddings, acting as a deterrent against codebook collapse, leading to high codebook utilization. Compared to VQ-VAE, our method obtains lower reconstruction errors under the same training conditions, trains in a fraction of the time, and with a constant number of parameters (equal to the embedding dimension $D$), making it a very scalable approach. We demonstrate these results on the FFHQ-1024 dataset and include FashionMNIST and Celeb-A.
BibTeX
@misc{khalil2023ll,
  author = {Ahmed Khalil and Robert Piechocki and Raul Santos-Rodriguez},
  title = {{LL-VQ-VAE: Learnable Lattice Vector-Quantization For Efficient Representations}},
  year = {2023},
  eprint = {2310.09382},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2310.09382},
}
Figure