Derandomizing Dense Binary Hypervector Codebooks for Quantized Scalars
Abstract
Hyperdimensional computing and vector symbolic architectures often represent quantized scalar levels by dense binary codebooks whose level-to-level similarity is intended to follow a prescribed function of scalar separation. At finite dimensionality, randomized scalar codebook constructions deviate from this target because of sampling noise, random-start imbalance, update-count fluctuations, component dependence, and finite-capacity effects. We develop a transition-based derandomization framework for dense binary scalar codebooks across two target-similarity families, with similarity decaying exponentially or linearly with level separation. The framework separates the target similarity law, the derandomization variant, and the concrete generator construction, making explicit how initialization, selection, update, and capacity-handling mechanisms shape the induced similarity profile. We formalize derandomization variants that separately constrain initial Hamming weight, update-count variability, and update balance, thereby controlling distinct sources of finite-dimensional error. For each family and variant, we derive the induced mean similarity, identify realization-wise and mean target-matching regimes, and derive exact finite-dimensional expressions for bias, variance, and root-mean-square error. Simulations across dimensions, quantization ranges, reference scalar levels, and generator constructions validate the theory and show how each constraint removes or reduces a specific source of similarity mismatch. The results provide practical guidance for choosing scalar codebook generators that more closely match a desired similarity law under finite-dimensional and hardware-relevant constraints.
Related
- Multi-Bitwidth Quantization for LLMs Using Additive Codebooks
- Perfect Reconstruction Cannot Show Whether Two Codebooks on a VAE's Latent Space Agree
- Sparse power spectrum for passive scalars under Pierrehumbert flow
- Lyapunov Exponents for Passive Scalars in White-in-Time Flows
- A Lagrangian Approach to Uniqueness for Active Scalars in Generalized Mean Oscillation Spaces
BibTeX
@misc{rachkovskij2026derandomizing,
title = {{Derandomizing Dense Binary Hypervector Codebooks for Quantized Scalars}},
author = {Dmitri Rachkovskij and Evgeny Osipov and Olexander Volkov and Denis Kleyko and Vaclav Snasel},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.38471},
eprint = {2609.38471},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.38471},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - UNPB
TI - Derandomizing Dense Binary Hypervector Codebooks for Quantized Scalars
AU - Dmitri Rachkovskij
AU - Evgeny Osipov
AU - Olexander Volkov
AU - Denis Kleyko
AU - Vaclav Snasel
PY - 2026
DA - 2026/09/29
PB - arXiv
DO - 10.48550/arxiv.2609.38471
UR - https://arxiv.org/abs/2609.38471
AB - Hyperdimensional computing and vector symbolic architectures often represent quantized scalar levels by dense binary codebooks whose level-to-level similarity is intended to follow a prescribed function of scalar separation. At finite dimensionality, randomized scalar codebook constructions deviate from this target because of sampling noise, random-start imbalance, update-count fluctuations, component dependence, and finite-capacity effects. We develop a transition-based derandomization framework for dense binary scalar codebooks across two target-similarity families, with similarity decaying exponentially or linearly with level separation. The framework separates the target similarity law, the derandomization variant, and the concrete generator construction, making explicit how initialization, selection, update, and capacity-handling mechanisms shape the induced similarity profile. We formalize derandomization variants that separately constrain initial Hamming weight, update-count variability, and update balance, thereby controlling distinct sources of finite-dimensional error. For each family and variant, we derive the induced mean similarity, identify realization-wise and mean target-matching regimes, and derive exact finite-dimensional expressions for bias, variance, and root-mean-square error. Simulations across dimensions, quantization ranges, reference scalar levels, and generator constructions validate the theory and show how each constraint removes or reduces a specific source of similarity mismatch. The results provide practical guidance for choosing scalar codebook generators that more closely match a desired similarity law under finite-dimensional and hardware-relevant constraints.
ER -