{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,10]],"date-time":"2026-02-10T18:03:05Z","timestamp":1770746585991,"version":"3.49.0"},"reference-count":52,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,2,10]],"date-time":"2026-02-10T00:00:00Z","timestamp":1770681600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>\n                    Hyperdimensional Computing (HDC) is a neurally inspired computing paradigm that leverages lightweight, high-dimensional operations to emulate key brain functions. Recent advances in HDC have primarily targeted two domains:\n                    <jats:italic>learning<\/jats:italic>\n                    , where the goal is to extract and generalize patterns for tasks such as classification, and\n                    <jats:italic>cognitive computation<\/jats:italic>\n                    , which requires accurate information retrieval for human-like reasoning. Although state-of-the-art HDC methods achieve strong performance in both areas, they lack a principled understanding of the fundamentally different requirements imposed by learning vs. cognition. In particular, existing works provide limited guidance on designing encoding methods that generate optimal hyperdimensional representations for these distinct tasks. In this study, we proposed the first\n                    <jats:italic>universal hyperdimensional encoding method<\/jats:italic>\n                    that dynamically adapts to the needs of both learning and cognitive computation. Our approach is based on neural-symbolic techniques that assign random complex hypervectors to atomic bases (e.g., alphabet definitions) and then apply algebraic operations in the high-dimensional\n                    <jats:italic>hyperspace<\/jats:italic>\n                    to control the correlation structure among encoded data points. Through theoretical analysis, we show that learning tasks benefit from\n                    <jats:italic>correlated<\/jats:italic>\n                    representations to maximize memorization and generalization capacity, whereas cognitive tasks require\n                    <jats:italic>orthogonal, highly separable<\/jats:italic>\n                    representations to enable accurate decoding and reasoning. We further derived a separation metric that quantifies this trade-off and validated it empirically across image classification and decoding tasks. Our results demonstrate that tuning the encoder to increase correlation improves classification accuracy from 65% to 95%, while maximizing separation enhances decoding accuracy from 85% to 100%. These findings provide the first systematic framework for designing hyperdimensional encoders that unify learning and cognition under a single, theoretically grounded representation model.\n                  <\/jats:p>","DOI":"10.3389\/frai.2026.1690492","type":"journal-article","created":{"date-parts":[[2026,2,10]],"date-time":"2026-02-10T07:17:09Z","timestamp":1770707829000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Optimal hyperdimensional representation for learning and cognitive computation"],"prefix":"10.3389","volume":"9","author":[{"given":"Prathyush P.","family":"Poduval","sequence":"first","affiliation":[{"name":"Donald Bren School of Information and Computer Sciences (ICS), University of California","place":["Irvine, Irvine, CA, United States"]}]},{"given":"Hamza","family":"Errahmouni Barkam","sequence":"additional","affiliation":[{"name":"Donald Bren School of Information and Computer Sciences (ICS), University of California","place":["Irvine, Irvine, CA, United States"]}]},{"given":"Xiangjian","family":"Liu","sequence":"additional","affiliation":[{"name":"Donald Bren School of Information and Computer Sciences (ICS), University of California","place":["Irvine, Irvine, CA, United States"]}]},{"given":"Sanggeon","family":"Yun","sequence":"additional","affiliation":[{"name":"Donald Bren School of Information and Computer Sciences (ICS), University of California","place":["Irvine, Irvine, CA, United States"]}]},{"given":"Yang","family":"Ni","sequence":"additional","affiliation":[{"name":"Purdue University Northwest","place":["Hammond, IN, United States"]}]},{"given":"Zhuowen","family":"Zou","sequence":"additional","affiliation":[{"name":"Donald Bren School of Information and Computer Sciences (ICS), University of California","place":["Irvine, Irvine, CA, United States"]}]},{"given":"Nathaniel D.","family":"Bastian","sequence":"additional","affiliation":[{"name":"United States Military Academy","place":["West Point, NY, United States"]}]},{"given":"Mohsen","family":"Imani","sequence":"additional","affiliation":[{"name":"Donald Bren School of Information and Computer Sciences (ICS), University of California","place":["Irvine, Irvine, CA, United States"]}]}],"member":"1965","published-online":{"date-parts":[[2026,2,10]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1145\/3695053.3731095","article-title":"\u201cHpvm-hdc: a heterogeneous programming system for accelerating hyperdimensional computing,\u201d","author":"Arbore","year":"2025","journal-title":"Proceedings of the 52nd Annual International Symposium on Computer Architecture"},{"key":"B2","doi-asserted-by":"publisher","first-page":"2311","DOI":"10.1162\/neco_a_01331","article-title":"Resonator networks, 1: an efficient solution for factoring high-dimensional, distributed representations of data structures","volume":"32","author":"Frady","year":"2020","journal-title":"Neural Comput"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.1145\/3517343.3522597","article-title":"\u201dComputing on functions using randomized vector representations,\u201d","author":"Frady","year":"2022","journal-title":"Neuro-Inspired Computational Elements Conference (NICE)"},{"key":"B4","article-title":"\u201cMultiplicative binding, representation operators and analogy,\u201d","author":"Gayler","year":"1998","journal-title":"International Conference on Cognitive Science, Workshop Poster"},{"key":"B5","first-page":"7","article-title":"\u201dReghd: robust and efficient regression in hyper-dimensional learning system,\u201d","volume-title":"DAC","author":"Hern\u00e1ndez-Cano","year":"2021"},{"key":"B6","first-page":"356","article-title":"\u201cDual: acceleration of clustering algorithms using digital-based processing in-memory,\u201d","volume-title":"MICRO","author":"Imani","year":"2020"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1145\/3489517.3530653","article-title":"\u201cNeural computation for robust and holographic face detection,\u201d","author":"Imani","year":"2022","journal-title":"Proceedings of the 59th ACM\/IEEE Design Automation Conference"},{"key":"B8","doi-asserted-by":"publisher","first-page":"118","DOI":"10.3389\/fnins.2011.00118","article-title":"Frontiers in neuromorphic engineering","volume":"5","author":"Indiveri","year":"2011","journal-title":"Front. Neurosci"},{"key":"B9","volume-title":"Sparse Distributed Memory","author":"Kanerva","year":"1988"},{"key":"B10","first-page":"869","article-title":"\u201cBinary spatter-coding of ordered k-tuples,\u201d","volume-title":"ICANN'96, Proceedings of the International Conference on Artificial Neural Networks, volume 1112 of Lecture Notes in Computer Science","author":"Kanerva","year":"1996"},{"key":"B11","doi-asserted-by":"crossref","first-page":"387","DOI":"10.1007\/978-1-4471-1599-1_57","article-title":"\u201dEncoding structure in boolean space,\u201d","volume-title":"ICANN 98","author":"Kanerva","year":"1998"},{"key":"B12","doi-asserted-by":"publisher","first-page":"139","DOI":"10.1007\/s12559-009-9009-8","article-title":"Hyperdimensional computing: an introduction to computing in distributed representation with high-dimensional random vectors","volume":"1","author":"Kanerva","year":"2009","journal-title":"Cogn. Comput"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.1109\/TCASAI.2024.3481193","article-title":"Private and efficient learning with hyperdimensional computing","author":"Khaleghi","year":"2024","journal-title":"IEEE Trans. Circuits Syst. Artif. Intell"},{"key":"B14","first-page":"38","article-title":"\u201cEfficient human activity recognition using hyperdimensional computing,\u201d","volume-title":"Proceedings of the 8th International Conference on the Internet of Things","author":"Kim","year":"2018"},{"key":"B15","doi-asserted-by":"publisher","first-page":"30766","DOI":"10.1109\/ACCESS.2018.2840128","article-title":"Hyperdimensional computing in industrial systems: the use-case of distributed fault isolation in a power plant","volume":"6","author":"Kleyko","year":"2018","journal-title":"IEEE Access"},{"key":"B16","doi-asserted-by":"publisher","first-page":"39","DOI":"10.1016\/j.procs.2016.07.404","article-title":"Modification of holographic graph neuron using sparse distributed representations","volume":"88","author":"Kleyko","year":"2016","journal-title":"Proc. Comput. Sci"},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.1145\/3538531","article-title":"A survey on hyperdimensional computing aka vector symbolic architectures, part i: models and data transformations","author":"Kleyko","year":"2021","journal-title":"arXiv preprint arXiv:2111.06077"},{"key":"B18","doi-asserted-by":"publisher","first-page":"2278","DOI":"10.1109\/5.726791","article-title":"Gradient-based learning applied to document recognition","volume":"86","author":"LeCun","year":"1998","journal-title":"Proc. IEEE"},{"key":"B19","doi-asserted-by":"publisher","first-page":"1","DOI":"10.5555\/3546258.3546366","article-title":"Towards a unified analysis of random fourier features","volume":"22","author":"Li","year":"2021","journal-title":"J. Mach. Learn. Res"},{"key":"B20","article-title":"\u201cMicrosoft coco: common objects in context,\u201d","volume-title":"Computer Vision \u2013 ECCV 2014. ECCV 2014. Lecture Notes in Computer Science","author":"Lin","year":"2015"},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.1162\/jocn_a_01544","article-title":"Convolutional neural networks as a model of the visual system: past, present, and future","author":"Lindsay","year":"2020","journal-title":"J. Cogn. Neurosci"},{"key":"B22","doi-asserted-by":"crossref","DOI":"10.4324\/9781410611147","volume-title":"Detection Theory: A User's Guide, 2nd Edn","author":"Macmillan","year":"2004"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.1126\/scirobotics.aaw6736","article-title":"Learning sensorimotor control with neuromorphic sensors: toward hyperdimensional active perception","author":"Mitrokhin","year":"2019","journal-title":"Sci. Robot"},{"key":"B24","doi-asserted-by":"publisher","first-page":"63","DOI":"10.3389\/frobt.2020.00063","article-title":"Symbolic representation and learning with hyperdimensional computing","volume":"7","author":"Mitrokhin","year":"2020","journal-title":"Front. Robot. AI"},{"key":"B25","doi-asserted-by":"publisher","first-page":"54","DOI":"10.1038\/s41928-020-00510-8","article-title":"A wearable biosensing system with in-sensor adaptive machine learning for hand gesture recognition","volume":"4","author":"Moin","year":"2021","journal-title":"Nat. Electr"},{"key":"B26","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/DAC.2018.8465801","article-title":"\u201dPulp-hd: accelerating brain-inspired high-dimensional computing on a parallel ultra-low power platform,\u201d","volume-title":"2018 55th ACM\/ESDA\/IEEE Design Automation Conference (DAC)","author":"Montagna","year":"2018"},{"key":"B27","article-title":"\u201cHyperdimensional computing for text classification,\u201d","author":"Najafabadi","year":"2016","journal-title":"DATE"},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.1109\/ICCAD66269.2025.11240977","article-title":"Hydra: sot-cam based vector symbolic macro for hyperdimensional computing","author":"Nayan","year":"2025","journal-title":"arXiv preprint arXiv:2504.14020"},{"key":"B29","first-page":"30","article-title":"\u201cHolographic reduced representations: convolution algebra for compositional distributed representations,\u201d","author":"Plate","year":"1991","journal-title":"IJCAI"},{"key":"B30","doi-asserted-by":"publisher","first-page":"623","DOI":"10.1109\/72.377968","article-title":"Holographic reduced representations","volume":"6","author":"Plate","year":"1995","journal-title":"IEEE Trans. Neural Netw"},{"key":"B31","doi-asserted-by":"crossref","DOI":"10.1109\/NANOARCH53687.2021.9642237","article-title":"\u201cRobust in-memory computing with hyperdimensional stochastic representation,\u201d","volume-title":"2021 IEEE\/ACM International Symposium on Nanoscale Architectures (NANOARCH)","author":"Poduval","year":""},{"key":"B32","doi-asserted-by":"publisher","DOI":"10.1145\/3489517.3530659","article-title":"\u201dAdaptive neural recovery for highly robust brain-like representation,\u201d","author":"Poduval","year":"","journal-title":"Proceedings of the 59th ACM\/IEEE Design Automation Conference"},{"key":"B33","doi-asserted-by":"publisher","first-page":"757125","DOI":"10.3389\/fnins.2022.757125","article-title":"Graphd: graph-based hyperdimensional memorization for brain-like cognitive learning","volume":"16","author":"Poduval","year":"","journal-title":"Front. Neurosci"},{"key":"B34","first-page":"1195","article-title":"\u201cStochd: stochastic hyperdimensional system for efficient and robust learning from raw data,\u201d","volume-title":"2021 58th ACM\/IEEE Design Automation Conference (DAC)","author":"Poduval","year":""},{"key":"B35","first-page":"781","article-title":"\u201cCognitive correlative encoding for genome sequence matching in hyperdimensional system,\u201d","volume-title":"2021 58th ACM\/IEEE Design Automation Conference (DAC)","author":"Poduval","year":""},{"key":"B36","doi-asserted-by":"publisher","first-page":"313","DOI":"10.1007\/s10559-015-9723-z","article-title":"Formation of similarity-reflecting binary vectors with random binary projections","volume":"51","author":"Rachkovskij","year":"2015","journal-title":"Cybernet. Syst. Anal"},{"key":"B37","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1615\/JAutomatInf","article-title":"Sparse binary distributed encoding of scalars","volume":"37","author":"Rachkovskiy","year":"2005","journal-title":"J. Autom. Inform. Sci"},{"key":"B38","doi-asserted-by":"publisher","first-page":"123","DOI":"10.1109\/JPROC.2018.2871163","article-title":"Efficient biosignal processing using hyperdimensional computing: network templates for combined learning and classification of exg signals","volume":"107","author":"Rahimi","year":"2018","journal-title":"Proc. IEEE"},{"key":"B39","article-title":"\u201cRandom features for large-scale kernel machines,\u201d","author":"Rahimi","year":"2007","journal-title":"Advances in Neural Information Processing Systems (NeurIPS)"},{"key":"B40","doi-asserted-by":"publisher","first-page":"1878","DOI":"10.1109\/TNNLS.2015.2462721","article-title":"Sequence prediction with sparse distributed hyperdimensional coding applied to the analysis of mobile phone use patterns","volume":"27","author":"R\u00e4s\u00e4nen","year":"2015","journal-title":"IEEE Trans. Neural Netw. Learn. Syst"},{"key":"B41","doi-asserted-by":"publisher","first-page":"641","DOI":"10.1038\/s42256-024-00848-0","article-title":"Neuromorphic visual scene understanding with resonator networks","volume":"6","author":"Renner","year":"2024","journal-title":"Nat. Mach. Intell"},{"key":"B42","doi-asserted-by":"publisher","first-page":"4523","DOI":"10.1007\/s10462-021-10110-3","article-title":"A comparison of vector symbolic architectures","volume":"55","author":"Schlegel","year":"2022","journal-title":"Artif. Intell. Rev"},{"key":"B43","doi-asserted-by":"publisher","first-page":"1371518","DOI":"10.3389\/fdata.2024.1371518","article-title":"An encoding framework for binarized images using hyperdimensional computing","volume":"7","author":"Smets","year":"2024","journal-title":"Front. Big Data"},{"key":"B44","doi-asserted-by":"publisher","first-page":"e1012426","DOI":"10.1371\/journal.pcbi.1012426","article-title":"Hyperdimensional computing: a fast, robust, and interpretable paradigm for biological data","volume":"20","author":"Stock","year":"2024","journal-title":"PLoS Comput. Biol"},{"key":"B45","doi-asserted-by":"publisher","first-page":"5143","DOI":"10.1609\/aaai.v33i01.33015143","article-title":"Holographic factorization machines for recommendation","volume":"33","author":"Tay","year":"2019","journal-title":"AAAI"},{"key":"B46","doi-asserted-by":"crossref","first-page":"84","DOI":"10.1109\/ISVLSI51109.2021.00026","article-title":"\u201cSpamhd: memory-efficient text spam detection using brain-inspired hyperdimensional computing,\u201d","volume-title":"2021 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)","author":"Thapa","year":"2021"},{"key":"B47","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-025-11181-2","article-title":"Classification using hyperdimensional computing: a review and comparative analysis","author":"Verg\u00e9s","year":"2025","journal-title":"Artif. Intell. Rev"},{"key":"B48","doi-asserted-by":"publisher","first-page":"87","DOI":"10.1038\/s41597-024-02933-w","article-title":"A dataset of oracle characters for benchmarking machine learning algorithms","volume":"11","author":"Wang","year":"2024","journal-title":"Sci. Data"},{"key":"B49","doi-asserted-by":"publisher","first-page":"2400228","DOI":"10.2991\/978-94-6463-600-0","article-title":"Hypersense: hyperdimensional intelligent sensing for energy-efficient sparse data processing","volume":"6","author":"Yin","year":"2024","journal-title":"Adv. Intell. Syst"},{"key":"B50","doi-asserted-by":"crossref","first-page":"211","DOI":"10.1109\/ASAP52443.2021.00039","article-title":"\u201cAssessing robustness of hyperdimensional computing against errors in associative memory,\u201d","volume-title":"2021 IEEE 32nd International Conference on Application-specific Systems, Architectures and Processors (ASAP)","author":"Zhang","year":"2021"},{"key":"B51","doi-asserted-by":"publisher","first-page":"858329","DOI":"10.3389\/fnins.2022.858329","article-title":"Eventhd: robust and efficient hyperdimensional learning from neuromorphic sensors","volume":"16","author":"Zou","year":"","journal-title":"Front. Neurosci"},{"key":"B52","doi-asserted-by":"publisher","DOI":"10.1145\/3470496.3527422","author":"Zou","year":"","journal-title":"Proceedings of the 49th Annual International Symposium on Computer Architecture"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1690492\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,10]],"date-time":"2026-02-10T07:17:13Z","timestamp":1770707833000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1690492\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,10]]},"references-count":52,"alternative-id":["10.3389\/frai.2026.1690492"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1690492","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,10]]},"article-number":"1690492"}}