Glued lattices are better quantizers than $K_{12}$
Erik Agrell, Daniel Pook-Kolb, and Bruce Allen

TL;DR
This paper demonstrates that newly constructed glued lattices outperform the traditional Coxeter-Todd lattice $K_{12}$ as quantizers in 12 dimensions by achieving lower normalized second moments.
Contribution
It introduces a novel method of constructing 12-dimensional lattices through gluing, resulting in better quantization performance than the long-standing $K_{12}$ lattice.
Findings
New lattices have lower normalized second moments than $K_{12}$
Gluing 6-dimensional lattices is effective for improving quantization
Challenged the long-held belief of $K_{12}$ optimality
Abstract
40 years ago, Conway and Sloane proposed using the highly symmetrical Coxeter-Todd lattice for quantization, and estimated its second moment. Since then, all published lists identify as the best 12-dimensional lattice quantizer. Surprisingly, is not optimal: we construct two new 12-dimensional lattices with lower normalized second moments. The new lattices are obtained by gluing together 6-dimensional lattices.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAdvanced Data Compression Techniques · Medical Imaging Techniques and Applications
