aiqbench analysis
SupHDC introduces a novel inference paradigm for hyperdimensional computing that leverages superposition to process multiple queries simultaneously, significantly reducing computational overhead without sacrificing accuracy. This approach reframes high-dimensional redundancy as a resource for shared computation, offering measurable speedups even on edge hardware like the Raspberry Pi 5. The work advances machine learning efficiency by demonstrating how structural properties of HDC can be exploited for scalable inference tasks.
Superposed Inference for Hyperdimensional Computing
arXiv:2609.32320v1 Announce Type: new Abstract: Hyperdimensional computing (HDC) is attractive for efficient and robust learning, but conventional inference still encodes every query independently, repeatedly paying the cost of high-dimensional projection. We introduce SupHDC, a new inference paradigm that processes multiple queries through a shared encoding computation. SupHDC assigns lightweight random slot keys, superposes the keyed queries before encoding, and uses slot-specific classifiers to recover their individual predictions. A random-feature kernel view explains why exact recovery of each hypervector is unnecessary: inference only needs to preserve the class evidence that determines the prediction. Across ten datasets, SupHDC achieves 1.39x analytical speedup with no average accuracy loss, and up to 2.08x speedup with only a 2.67 percentage-point mean accuracy loss. On a Raspberry Pi~5, it delivers 2.01x measured wall-clock speedup with a 2.26 percentage-point loss in mean prediction accuracy. SupHDC shows that high-dimensional redundancy can be used not only for robustness, but also as capacity for shared inference.
Abstract:Hyperdimensional computing (HDC) is attractive for efficient and robust learning, but conventional inference still encodes every query independently, repeatedly paying the cost of high-dimensional projection. We introduce SupHDC, a new inference paradigm that processes multiple queries through a shared encoding computation. SupHDC assigns lightweight random slot keys, superposes the keyed queries before encoding, and uses slot-specific classifiers to recover their individual predictions. A random-feature kernel view explains why exact recovery of each hypervector is unnecessary: inference only needs to preserve the class evidence that determines the prediction. Across ten datasets, SupHDC achieves 1.39x analytical speedup with no average accuracy loss, and up to 2.08x speedup with only a 2.67 percentage-point mean accuracy loss. On a Raspberry Pi~5, it delivers 2.01x measured wall-clock speedup with a 2.26 percentage-point loss in mean prediction accuracy. SupHDC shows that high-dimensional redundancy can be used not only for robustness, but also as capacity for shared inference.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2609.32320 [cs.LG]
(or arXiv:2609.32320v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2609.32320 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 248 words
Related coverage
AMS-HD: Hyperdimensional Computing for Real-Time and Energy-Efficient Acute Mountain Sickness Detection
arXiv · Jun 9, 2026
BiHDTrans: binary hyperdimensional transformer for efficient multivariate time series classification
arXiv · Sep 11, 2026
Logit-Aware MIMO AirComp for Distributed Mixture-of-Experts LLM Inference over Wireless Edge Networks
arXiv · Oct 7, 2026
Privacy-Preserving Decentralized Federated Learning via Explainable Adaptive Differential Privacy
arXiv · Aug 18, 2026