Learning from Hypervectors: A Survey on Hyperdimensional Encoding
Hyperdimensional computing (HDC) is an emerging computing paradigm that imitates the brain’s structure to offer a powerful and efficient processing and learning model. In HDC, the data are encoded with long vectors, called hypervectors , which typically range from 1, 000 to 10, 000 in length. The literature provides several encoding methods to generate orthogonal or correlated hypervectors, based on the application. Existing surveys focus on overall HDC system aspects like inputs, primary computations, and outputs. However, this study takes a more specific approach. It zeroes in on the HDC system input and the generation of hypervectors, directly influencing the hypervector encoding. This survey brings together various hypervector generation and encoding methods, exploring the limitations, challenges, and potential benefits they entail.
Authors
- M. Hassan Najafi (ORCID: https://orcid.org/0000-0002-4655-6229)
- Mohsen Imani (ORCID: https://orcid.org/0000-0002-5761-0622)
- Sercan Aygün (ORCID: https://orcid.org/0000-0002-4615-7914)
- Mehran Shoushtari Moghadam (ORCID: https://orcid.org/0000-0002-1325-1664)
Institutions
- University of California, Irvine (US)
- Case Western Reserve University (US)
- University of Louisiana at Lafayette (US)
Publication Details
- Journal
- ACM Computing Surveys
- Published
- 2026-11-19
- DOI
- https://doi.org/10.1145/3863975
- Citations
- 68
- Primary Topic
- Cognitive Computing and Networks
- Type
- article
- Field-Weighted Citation Impact
- 18.39
Funders
- National Science Foundation
- Semiconductor Research Corporation
- Cisco Systems
- Nvidia
- U.S. Navy
- Office of Naval Research