# DecoHD: Decomposed Hyperdimensional Classification under Extreme Memory Budgets

arXiv: https://arxiv.org/abs/2511.03911 · HTML: https://mlsanity.com/paper/2511.03911/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 41 references that are in the map.
- First posted on arXiv 5 Nov 2025.

## Builds on (most cited first, 12 of 12)

- [Classification using Hyperdimensional Computing: A Review](https://mlsanity.com/paper/2004.11204/) (2020)
- [Towards Efficient Tensor Decomposition-Based DNN Model Compression with Optimization Framework](https://mlsanity.com/paper/2107.12422/) (2021)
- [TinyTrain: Resource-Aware Task-Adaptive Sparse Training of DNNs at the Data-Scarce Edge](https://mlsanity.com/paper/2307.09988/) (2023)
- [LoFT: Low-Rank Adaptation That Behaves Like Full Fine-Tuning](https://mlsanity.com/paper/2505.21289/) (2025)
- [DistHD: A Learner-Aware Dynamic Encoding Method for Hyperdimensional Classification](https://mlsanity.com/paper/2304.05503/) (2023)
- [The Future of Consumer Edge-AI Computing](https://mlsanity.com/paper/2210.10514/) (2022)
- [Maestro: Uncovering Low-Rank Structures via Trainable Decomposition](https://mlsanity.com/paper/2308.14929/) (2023)
- [HyperSense: Hyperdimensional Intelligent Sensing for Energy-Efficient Sparse Data Processing](https://mlsanity.com/paper/2401.10267/) (2024)
- [NeuroHash: A Hyperdimensional Neuro-Symbolic Framework for Spatially-Aware Image Hashing and Retrieval](https://mlsanity.com/paper/2404.11025/) (2024)
- [FedorAS: Federated Architecture Search under system heterogeneity](https://mlsanity.com/paper/2206.11239/) (2022)
- [PriPrune: Quantifying and Preserving Privacy in Pruned Federated Learning](https://mlsanity.com/paper/2310.19958/) (2023)
- [Hyperdimensional Intelligent Sensing for Efficient Real-Time Audio Processing on Extreme Edge](https://mlsanity.com/paper/2502.10718/) (2025)

## Built on by (most cited first, 2 of 2)

- [FusionSense: Tri-Stage Near-Sensor Learning for Runtime-Adaptive Multimodal Edge Intelligence](https://mlsanity.com/paper/2605.22868/) (2026)
- [Qubit-Efficient Quantum Search for Hyperdimensional Decomposition via Logarithmic Encoding](https://mlsanity.com/paper/2607.11936/) (2026)

## Abstract (arXiv)

Decomposition is a proven way to shrink deep networks without changing input-output dimensionality or interface semantics. We bring this idea to hyperdimensional computing (HDC), where footprint cuts usually shrink the feature axis and erode concentration and robustness. Prior HDC decompositions decode via fixed atomic hypervectors, which are ill-suited for compressing learned class prototypes. We introduce DecoHD, which learns directly in a decomposed HDC parameterization: a small, shared set of per-layer channels with multiplicative binding across layers and bundling at the end, yielding a large representational space from compact factors. DecoHD compresses along the class axis via a lightweight bundling head while preserving native bind-bundle-score; training is end-to-end, and inference remains pure HDC, aligning with in/near-memory accelerators. In evaluation, DecoHD attains extreme memory savings with only minor accuracy degradation under tight deployment budgets. On average it stays within about 0.1-0.15% of a strong non-reduced HDC baseline (worst case 5.7%), is more robust to random bit-flip noise, reaches its accuracy plateau with up to ~97% fewer trainable parameters, and--in hardware--delivers roughly 277x/35x energy/speed gains over a CPU (AMD Ryzen 9 9950X), 13.5x/3.7x over a GPU (NVIDIA RTX 4090), and 2.0x/2.4x over a baseline HDC ASIC.